Simulace.info - User contributions [en]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T21:04:12Z

Doum01:

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

In Table 1 you can find the descriptive statistics of the outputted demographics data. 5th and 95th percentile are added so we can visually see the 90% confidence interval. To not repeat the obvious multiple times, I shall describe the meaning of the statistics only for Czech language tests for students of 4-year long high schools. There is a 90% likelihood that between 203253 and 203367 people will be in the age group that is able to apply for the high school. With the same likelihood 67057 and 72125 will actually apply for Cermat high school exams. There is 90% likelihood that the number of students that will write the Czech test will be anywhere between 64892 and 70123. All of the other data is read the same way.

[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]

These demographics were applied to the simulated distribution as can be seen in Figure 2. It is clear that the median distribution is of a nearly same shape as our baseline distribution of summed values, this is an expected behaviour. The detailed data for these graphs are included in the output files. The number are not viable to be shown in nongraphical way in this wikipage. It is derived from the raw outputted data with a simple Excel function.

[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]

= Conclusion =

The demographic part of this model can be used by Cermat to see how many applicants they can expect, the model in its entirety can be used either for high school to know how many students to prepare for and how will the point limit for their school will most likely look like or a students who would like to know how do they rank in country on average before the Cermat's annual report gets made many months later.

The best possible improvements for the model would lie in more accurate population estimates based on the detailed historical data of population, mortality and immigration. More advanced statistical methods could be used for the fitting the distribution which would require a more advanced statistical software (R,Python,Stata). Transformation of the model to a different program could also drastically improve the effiency and computation limits.

During the process of the model's creation it has become appearant that the model might not be immediatly useful as if the Covid-19 situation does not improve, Cermat tests will not be mandatory where they are mandatory now. That would turn the entire part of demographic data for applicants on its head, and would have a highly significant effect, that is impossible to quantify before the final decision is made by both the governement and the high schools themselves.

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

= Xlsb model and output file=
[[File:Cermat_model.zip]]

File:Cermat model.zip

2021-01-20T21:03:03Z

Doum01:

WS 2020/2021

2021-01-20T20:58:54Z

Doum01:

Semestral papers from winter term 2020/2021. Please, put here links to the pages with your paper. First you need to have your [[Assignments WS 2020/2021|assignment approved]]

[http://www.simulace.info/index.php/Impact_of_late_lockdown_and_hospital_capacity_on_19-COVID_spread Impact of late lockdown and hospital capacity on 19-COVID spread] , Thomas BAEUMLIN [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 16:15, 7 January 2020 (CET)

[[Comparing_multiple_airplane_boarding_methods]], Adam Jedlička [[User:Jeda00|jeda00]] 20:45, 17 January 2020 (CET)

[http://www.simulace.info/index.php/Effects_of_VAT_exemption_abolition_to_imported_low_value_consignments Effects of VAT exemption abolition to imported low value consignments], Adela Barchankova [[User:bara15|bara15]] ([[User talk:bara15|talk]]) 21:00, 18 January 2020 (CET)

[[Retirement Planning]], [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 02:08, 19 January 2021 (CET)

[[Covid 19 - Contacts]], [[User:praa03|praa03]] 17:08, 19 January 2021 (CET)

[[Corrupted Blood Incident]], Michal Frýba, [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 19:32, 19 January 2021 (CET)

[[Car acquisition]], [[Mico00|Mico00]] 19:59, 19 January 2021 (CET)

[[Widening the spread between rich and poor]], [[User:Sára|Sára]] 21:40, 19 January 2021 (CET)

[[Development of real estate prices in Prague]], [[User:netp02|netp02]] 23:41, 19 January 2021 (CET)

[[IT Team simulation]], [[Antonsverlov]] , 20 January 2021 (CET)

[[Evolution of retirement pay in the Czech Republic]], Arnošt Sychra [[User:Syca00|Syca00]] 03:30, 20 January 2020 (CET)

[http://www.simulace.info/index.php/COVID-19_symptoms COVID-19 symptoms simulation] , Kamila Steinbartová [[User:stek15|stek15]] ([[User talk:stek15|talk]]) 13:07, 20 January 2020 (CET)

[http://www.simulace.info/index.php/Risk_Simulation_in_Banking Risk Simulation in Banking], Simon Orlik [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 16:52, 20 January (CET)

[http://www.simulace.info/index.php/Garbage_collection Garbage collection] [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 19:32, 20 January 2021 (CET)

[[Covid-19 Mass testing]], Milan Pavelčák [[User:pavm13|pavm13]] 20:50, 20 January 2021 (CET)

[http://www.simulace.info/index.php/Efficient_Recruitment_Process Efficient Recruitment Process] [[[User:Gautier Dubois|Gautier Dubois]] ([[User talk:Gautier Dubois|talk]]) 20:51, 20 January 2021 (CET)]

[[Simulation optimalization of incident solving]], [[User:Glyn00|Glyn00]] 21:18, 20 January 2020 (CET)

[[Predicting number of applicants for high school exams and their level of success in 2021]], Matyáš Douda [[User:MD|MD]] 21:54, 20 January 2020 (CET)

WS 2020/2021

2021-01-20T20:57:38Z

Doum01:

Semestral papers from winter term 2020/2021. Please, put here links to the pages with your paper. First you need to have your [[Assignments WS 2020/2021|assignment approved]]

[http://www.simulace.info/index.php/Impact_of_late_lockdown_and_hospital_capacity_on_19-COVID_spread Impact of late lockdown and hospital capacity on 19-COVID spread] , Thomas BAEUMLIN [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 16:15, 7 January 2020 (CET)

[[Comparing_multiple_airplane_boarding_methods]], Adam Jedlička [[User:Jeda00|jeda00]] 20:45, 17 January 2020 (CET)

[http://www.simulace.info/index.php/Effects_of_VAT_exemption_abolition_to_imported_low_value_consignments Effects of VAT exemption abolition to imported low value consignments], Adela Barchankova [[User:bara15|bara15]] ([[User talk:bara15|talk]]) 21:00, 18 January 2020 (CET)

[[Retirement Planning]], [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 02:08, 19 January 2021 (CET)

[[Covid 19 - Contacts]], [[User:praa03|praa03]] 17:08, 19 January 2021 (CET)

[[Corrupted Blood Incident]], Michal Frýba, [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 19:32, 19 January 2021 (CET)

[[Car acquisition]], [[Mico00|Mico00]] 19:59, 19 January 2021 (CET)

[[Widening the spread between rich and poor]], [[User:Sára|Sára]] 21:40, 19 January 2021 (CET)

[[Development of real estate prices in Prague]], [[User:netp02|netp02]] 23:41, 19 January 2021 (CET)

[[IT Team simulation]], [[Antonsverlov]] , 20 January 2021 (CET)

[[Evolution of retirement pay in the Czech Republic]], Arnošt Sychra [[User:Syca00|Syca00]] 03:30, 20 January 2020 (CET)

[http://www.simulace.info/index.php/COVID-19_symptoms COVID-19 symptoms simulation] , Kamila Steinbartová [[User:stek15|stek15]] ([[User talk:stek15|talk]]) 13:07, 20 January 2020 (CET)

[http://www.simulace.info/index.php/Risk_Simulation_in_Banking Risk Simulation in Banking], Simon Orlik [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 16:52, 20 January (CET)

[http://www.simulace.info/index.php/Garbage_collection Garbage collection] [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 19:32, 20 January 2021 (CET)

[[Covid-19 Mass testing]], Milan Pavelčák [[User:pavm13|pavm13]] 20:50, 20 January 2021 (CET)

[http://www.simulace.info/index.php/Efficient_Recruitment_Process Efficient Recruitment Process] [[[User:Gautier Dubois|Gautier Dubois]] ([[User talk:Gautier Dubois|talk]]) 20:51, 20 January 2021 (CET)]

[[Predicting number of applicants for high school exams and their level of success in 2021]], Matyáš Douda [[User:MD|MD]] 21:54, 20 January 2020 (CET)

[[Simulation optimalization of incident solving]], [[User:Glyn00|Glyn00]] 21:18, 20 January 2020 (CET)

WS 2020/2021

2021-01-20T20:56:52Z

Doum01: Undo revision 20972 by Doum01 (talk)

Semestral papers from winter term 2020/2021. Please, put here links to the pages with your paper. First you need to have your [[Assignments WS 2020/2021|assignment approved]]

[http://www.simulace.info/index.php/Impact_of_late_lockdown_and_hospital_capacity_on_19-COVID_spread Impact of late lockdown and hospital capacity on 19-COVID spread] , Thomas BAEUMLIN [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 16:15, 7 January 2020 (CET)

[[Comparing_multiple_airplane_boarding_methods]], Adam Jedlička [[User:Jeda00|jeda00]] 20:45, 17 January 2020 (CET)

[http://www.simulace.info/index.php/Effects_of_VAT_exemption_abolition_to_imported_low_value_consignments Effects of VAT exemption abolition to imported low value consignments], Adela Barchankova [[User:bara15|bara15]] ([[User talk:bara15|talk]]) 21:00, 18 January 2020 (CET)

[[Retirement Planning]], [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 02:08, 19 January 2021 (CET)

[[Covid 19 - Contacts]], [[User:praa03|praa03]] 17:08, 19 January 2021 (CET)

[[Corrupted Blood Incident]], Michal Frýba, [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 19:32, 19 January 2021 (CET)

[[Car acquisition]], [[Mico00|Mico00]] 19:59, 19 January 2021 (CET)

[[Widening the spread between rich and poor]], [[User:Sára|Sára]] 21:40, 19 January 2021 (CET)

[[Development of real estate prices in Prague]], [[User:netp02|netp02]] 23:41, 19 January 2021 (CET)

[[IT Team simulation]], [[Antonsverlov]] , 20 January 2021 (CET)

[[Evolution of retirement pay in the Czech Republic]], Arnošt Sychra [[User:Syca00|Syca00]] 03:30, 20 January 2020 (CET)

[http://www.simulace.info/index.php/COVID-19_symptoms COVID-19 symptoms simulation] , Kamila Steinbartová [[User:stek15|stek15]] ([[User talk:stek15|talk]]) 13:07, 20 January 2020 (CET)

[http://www.simulace.info/index.php/Risk_Simulation_in_Banking Risk Simulation in Banking], Simon Orlik [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 16:52, 20 January (CET)

[http://www.simulace.info/index.php/Garbage_collection Garbage collection] [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 19:32, 20 January 2021 (CET)

[[Covid-19 Mass testing]], Milan Pavelčák [[User:pavm13|pavm13]] 20:50, 20 January 2021 (CET)

[http://www.simulace.info/index.php/Efficient_Recruitment_Process Efficient Recruitment Process] [[[User:Gautier Dubois|Gautier Dubois]] ([[User talk:Gautier Dubois|talk]]) 20:51, 20 January 2021 (CET)]

[http://www.simulace.info/index.php/Predicting_number_of_applicants_for_high_school_exams_and_their_level_of_success_in_2021#Cermat_data_analysis], Matyáš Douda [[User:MD|MD]] 21:54, 20 January 2020 (CET)

[[Simulation optimalization of incident solving]], [[User:Glyn00|Glyn00]] 21:18, 20 January 2020 (CET)

WS 2020/2021

2021-01-20T20:56:15Z

Doum01:

Semestral papers from winter term 2020/2021. Please, put here links to the pages with your paper. First you need to have your [[Assignments WS 2020/2021|assignment approved]]

[http://www.simulace.info/index.php/Impact_of_late_lockdown_and_hospital_capacity_on_19-COVID_spread Impact of late lockdown and hospital capacity on 19-COVID spread] , Thomas BAEUMLIN [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 16:15, 7 January 2020 (CET)

[[Comparing_multiple_airplane_boarding_methods]], Adam Jedlička [[User:Jeda00|jeda00]] 20:45, 17 January 2020 (CET)

[http://www.simulace.info/index.php/Effects_of_VAT_exemption_abolition_to_imported_low_value_consignments Effects of VAT exemption abolition to imported low value consignments], Adela Barchankova [[User:bara15|bara15]] ([[User talk:bara15|talk]]) 21:00, 18 January 2020 (CET)

[[Retirement Planning]], [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 02:08, 19 January 2021 (CET)

[[Covid 19 - Contacts]], [[User:praa03|praa03]] 17:08, 19 January 2021 (CET)

[[Corrupted Blood Incident]], Michal Frýba, [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 19:32, 19 January 2021 (CET)

[[Car acquisition]], [[Mico00|Mico00]] 19:59, 19 January 2021 (CET)

[[Widening the spread between rich and poor]], [[User:Sára|Sára]] 21:40, 19 January 2021 (CET)

[[Development of real estate prices in Prague]], [[User:netp02|netp02]] 23:41, 19 January 2021 (CET)

[[IT Team simulation]], [[Antonsverlov]] , 20 January 2021 (CET)

[[Evolution of retirement pay in the Czech Republic]], Arnošt Sychra [[User:Syca00|Syca00]] 03:30, 20 January 2020 (CET)

[http://www.simulace.info/index.php/COVID-19_symptoms COVID-19 symptoms simulation] , Kamila Steinbartová [[User:stek15|stek15]] ([[User talk:stek15|talk]]) 13:07, 20 January 2020 (CET)

[http://www.simulace.info/index.php/Risk_Simulation_in_Banking Risk Simulation in Banking], Simon Orlik [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 16:52, 20 January (CET)

[http://www.simulace.info/index.php/Garbage_collection Garbage collection] [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 19:32, 20 January 2021 (CET)

[[Covid-19 Mass testing]], Milan Pavelčák [[User:pavm13|pavm13]] 20:50, 20 January 2021 (CET)

[http://www.simulace.info/index.php/Efficient_Recruitment_Process Efficient Recruitment Process] [[[User:Gautier Dubois|Gautier Dubois]] ([[User talk:Gautier Dubois|talk]]) 20:51, 20 January 2021 (CET)]

[http://www.simulace.info/index.php/Predicting_number_of_applicants_for_high_school_exams_and_their_level_of_success_in_2021], Matyáš Douda [[User:MD|MD]] 21:54, 20 January 2020 (CET)

[[Simulation optimalization of incident solving]], [[User:Glyn00|Glyn00]] 21:18, 20 January 2020 (CET)

WS 2020/2021

2021-01-20T20:55:05Z

Doum01:

Semestral papers from winter term 2020/2021. Please, put here links to the pages with your paper. First you need to have your [[Assignments WS 2020/2021|assignment approved]]

[http://www.simulace.info/index.php/Impact_of_late_lockdown_and_hospital_capacity_on_19-COVID_spread Impact of late lockdown and hospital capacity on 19-COVID spread] , Thomas BAEUMLIN [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 16:15, 7 January 2020 (CET)

[[Comparing_multiple_airplane_boarding_methods]], Adam Jedlička [[User:Jeda00|jeda00]] 20:45, 17 January 2020 (CET)

[http://www.simulace.info/index.php/Effects_of_VAT_exemption_abolition_to_imported_low_value_consignments Effects of VAT exemption abolition to imported low value consignments], Adela Barchankova [[User:bara15|bara15]] ([[User talk:bara15|talk]]) 21:00, 18 January 2020 (CET)

[[Retirement Planning]], [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 02:08, 19 January 2021 (CET)

[[Covid 19 - Contacts]], [[User:praa03|praa03]] 17:08, 19 January 2021 (CET)

[[Corrupted Blood Incident]], Michal Frýba, [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 19:32, 19 January 2021 (CET)

[[Car acquisition]], [[Mico00|Mico00]] 19:59, 19 January 2021 (CET)

[[Widening the spread between rich and poor]], [[User:Sára|Sára]] 21:40, 19 January 2021 (CET)

[[Development of real estate prices in Prague]], [[User:netp02|netp02]] 23:41, 19 January 2021 (CET)

[[IT Team simulation]], [[Antonsverlov]] , 20 January 2021 (CET)

[[Evolution of retirement pay in the Czech Republic]], Arnošt Sychra [[User:Syca00|Syca00]] 03:30, 20 January 2020 (CET)

[http://www.simulace.info/index.php/COVID-19_symptoms COVID-19 symptoms simulation] , Kamila Steinbartová [[User:stek15|stek15]] ([[User talk:stek15|talk]]) 13:07, 20 January 2020 (CET)

[http://www.simulace.info/index.php/Risk_Simulation_in_Banking Risk Simulation in Banking], Simon Orlik [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 16:52, 20 January (CET)

[http://www.simulace.info/index.php/Garbage_collection Garbage collection] [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 19:32, 20 January 2021 (CET)

[[Covid-19 Mass testing]], Milan Pavelčák [[User:pavm13|pavm13]] 20:50, 20 January 2021 (CET)

[http://www.simulace.info/index.php/Efficient_Recruitment_Process Efficient Recruitment Process] [[[User:Gautier Dubois|Gautier Dubois]] ([[User talk:Gautier Dubois|talk]]) 20:51, 20 January 2021 (CET)]

[http://www.simulace.info/index.php/Predicting_number_of_applicants_for_high_school_exams_and_their_level_of_success_in_2021#Cermat_data_analysis], Matyáš Douda [[User:MD|MD]] 21:54, 20 January 2020 (CET)

[[Simulation optimalization of incident solving]], [[User:Glyn00|Glyn00]] 21:18, 20 January 2020 (CET)

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:52:31Z

Doum01: /* Conclusion */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:42:28Z

Doum01: /* Results */

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

In Table 1 you can find the descriptive statistics of the outputted demographics data. 5th and 95th percentile are added so we can visually see the 90% confidence interval. To not repeat the obvious multiple times, I shall describe the meaning of the statistics only for Czech language tests for students of 4-year long high schools. There is a 90% likelihood that between 203253 and 203367 people will be in the age group that is able to apply for the high school. With the same likelihood 67057 and 72125 will actually apply for Cermat high school exams. There is 90% likelihood that the number of students that will write the Czech test will be anywhere between 64892 and 70123. All of the other data is read the same way.

[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]

These demographics were applied to the simulated distribution as can be seen in Figure 2. It is clear that the median distribution is of a nearly same shape as our baseline distribution of summed values, this is an expected behaviour. The detailed data for these graphs are included in the output files. The number are not viable to be shown in nongraphical way in this wikipage. It is derived from the raw outputted data with a simple Excel function.

[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:39:10Z

Doum01: /* Results */

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

In Table 1 you can find the descriptive statistics of the outputted demographics data. 5th and 95th percentile are added so we can visually see the 90% confidence interval. To not repeat the obvious multiple times, I shall describe the meaning of the statistics only for Czech language tests for students of 4-year long high schools. There is a 90% likelihood that between 203253 and 203367 people will be in the age group that is able to apply for the high school. With the same likelihood 67057 and 72125 will actually apply for Cermat high school exams. There is 90% likelihood that the number of students that will write the Czech test will be anywhere between 64892 and 70123. All of the other data is read the same way.

[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]

These demographics were applied to the simulated distribution as can be seen in Figure 2. It is clear that the median distribution is of a nearly same shape as our baseline distribution of summed values, this is an expected behaviour.

[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:36:05Z

Doum01: /* Results */

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

In Table 1 you can find the descriptive statistics of the outputted demographics data. 5th and 95th percentile are added so we can visually see the 90% confidence interval. To not repeat the obvious multiple times, I shall describe the meaning of the statistics only for Czech language tests for students of 4-year long high schools. There is a 90% likelihood that between 203253 and 203367 people will be in the age group that is able to apply for the high school. With the same likelihood 67057 and 72125 will actually apply for Cermat high school exams. There is 90% likelihood that the number of students that will write the Czech test will be anywhere between 64892 and 70123. All of the other data is read the same way.

[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]

These demographics were applied to the simulated distribution as can be seen in Figure 2.

[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:29:47Z

Doum01: /* Results */

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

In Table 1 you can find the descriptive statistics of the outputted demographics data. 5th and 95th percentile are added so we can visually see the 90% confidence interval.
[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]
[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:25:56Z

Doum01: /* Results */

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]
[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:25:03Z

Doum01: /* Source */

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]
[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 CZSO Annual Reports - 3 more for preceding years were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Cermat - Raw data]
#[https://prijimacky.cermat.cz/ Cermat main page - Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:24:12Z

Doum01:

__TOC__

= Problem Definition =
This model serves to predict the population, number of applicants and number of the various kinds of tests that are written in the Czech generalised national high school entry exam, which is mandatory for all high school that end with the final "Maturita exam". As any prediction model drastically loses accuracy for the further years, it is designed in the way that it can be easily released for year 2022 when the real data for 2021 is available.

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables. The model uses the .xlsb format for the better optimazation of file size and efficiency.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

[[File:Figure1.JPG|center|thumb|480 px|Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom]]

= Results =

[[File:Figure2.JPG|center|thumb|900 px|Figure 2 - 90% confidence intervals of the simulated distributions]]
[[File:Table1.JPG|center|thumb|720 px|Table 1 - Simulated values]]

= Conclusion =

= Source =
#[https://www.czso.cz/csu/czso/statisticka-rocenka-ceske-republiky-2020 3 more annual reports were used]
#[https://vysledky.cermat.cz/statistika/Default.aspx Raw data]
#[https://prijimacky.cermat.cz/ Basic information about present situation and the future]

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:17:44Z

Doum01: /* Results */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:17:19Z

Doum01: /* Results */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:16:58Z

Doum01: /* Results */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:15:11Z

Doum01: /* Cermat data analysis */

File:Table1.JPG

2021-01-20T20:13:56Z

Doum01: Table 1 - Simulated values

== Summary ==
Table 1 - Simulated values

File:Figure2.JPG

2021-01-20T20:13:21Z

Doum01: Figure 2 - 90% confidence intervals of the estimated distributions

== Summary ==
Figure 2 - 90% confidence intervals of the estimated distributions

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:12:21Z

Doum01: /* Method */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:11:16Z

Doum01: /* Cermat data analysis */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:10:12Z

Doum01: /* Cermat data analysis */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:09:37Z

Doum01: /* Cermat data analysis */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:09:13Z

Doum01: /* Cermat data analysis */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:08:59Z

Doum01: /* Cermat data analysis */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:08:32Z

Doum01:

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T20:08:05Z

Doum01:

File:Figure1.JPG

2021-01-20T20:04:38Z

Doum01: Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom

== Summary ==
Figure 1 - Distribution of the students in summed data on the top, distribution of the students in the current iteration of the simulation on the bottom

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T19:57:48Z

Doum01:

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T19:57:22Z

Doum01: /* Problem Definition */

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T19:52:34Z

Doum01: /* Method */

__TOC__

= Problem Definition =

= Method =
The method for this work is Monte Carlo application in Excel, VBA is used as a more controlled substitute to Data Tables.

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T19:51:09Z

Doum01:

__TOC__

= Problem Definition =

= Method =

= Model =

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

== Demographics ==

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

== Cermat data analysis ==

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T19:49:53Z

Doum01: /* Model */

__TOC__

== Definice problému ==

== Method ==

== Model ==

The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.

= Demographics =

The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.

The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.

Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.

Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.

The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

= Cermat data analysis =

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

Predicting number of applicants for high school exams and their level of success in 2021

2021-01-20T19:48:32Z

Doum01: Created page with " __TOC__ == Definice problému == == Method == == Model == The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from..."

__TOC__

== Definice problému ==

== Method ==

== Model ==
The model consists of 2 parts, one of them uses data from Czech Statistical Office(CZSO) and annual reports from Cermat to estimate the population, number of applicants and how many applicants will write which tests. The second part applies data from the first part on the distributions that we got from raw data from Cermat.
The first part starts with 3 input tables from Cermat, the first one is the total number of applicants which we got from the the annual reports. It is expected that in the years where there were 2 rounds (2017-2019), only students that took part in the first round can take part in the second. This would match the absolute growth between 2019 and 2020 that was reported in Cermat’s latest press release as well. It’s worth noting that the data for 8 year High Schools is significantly higher in 2019 than in other years. The other two tables sum all tests taken that are present in the raw data.
The next table consists of CZSO’s yearly reports – population by age tables. Important things to notice in the table are the study type that is assigned to age on left (for the 4 year, 6 year and 8 year high school) and the fact that each generation gains number of people each year.
Because of that we are looking up the growths in the next table. We can generalize that the rate of growth is usually the same or higher. Therefore the rate of growth is set as normal distribution with the expected value same as the last growth traced, with standard error based on the data we have and probability limited to the top 50% of cases. As that means we cannot generate the value with the RAND() function, we’re replacing it with a more complex RANDOM() function. The last column in this table adds the estimated growth to the generations.
Population and applicants in history tables shows historical population for each type of study, number of applicants is added again for clarity. This serves as a base for calculation of shares between population and applicants throughout the years.
The next table calculates these shares with simple division, the data for 2021 is then predicted with normal distribution.
The shares between applicants and tests looks at the shares of them, while 2021 prediction uses the same logic, the expected value for the normal distribution is much higher. As such the values can exceed realms of possibility, which has been counteracted by capping the shares to 99.5%
The final table shows the forecasted values that are used in the second part of the model. Population is calculated by simple addition, the rest is the product of the estimated share and population and applicants respectively.

Raw data from Cermat includes all answers and points achieved for every question, it is also separated on sheets by each test term (1st Official term, 1st Substitute term, 2nd Official term, 2nd Substitute term – described as A,B,C,D respectively) The point gains have been summed and exported to the excel file. The data has then been transformed to consist of number of students that achieved each point count. For less bias Official terms have been combined with their Substitute terms, as the Substitute terms always have only a fraction of the students that took the Official ones. To find the distributions for points for each study type and subject, the data from all of the test through the years has been combined and plotted (Figure 1). From the data we estimate Normal distribution for Czech language and Log – Normal distribution for the Mathematics. Statistical analysis in a different software could estimate a more accurate distribution for the Mathematics.

To avoid the effect of numbers of applicant’s in the distribution the data has been changed to shares. The shares from the combined and plotted data has been used as the expected value, standard deviation of the preceding data is used as a standard deviation for the normal distribution. We are basically treating each possible point gain value as a different normal distribution, this allows us to avoid problems with unfitting assumed distributions. The sum of the new shares is obviously not 1 – which does not make sense for a probability distribution. The values are therefore corrected and adjusted, values less than 0 are turned to 0 and then they are divided by the adjustment factor, which is calculated as the sum of the unadjusted values.

Finally the number of students for each point gain value is calculated from the number of previously forecasted tests and the shares that we have generated in this point. These number are also plotted and compared with their matching source distributions for a sanity check (Figure 1). As the numbers have to be rounded, we end up with a new forecast for the number of tests taken. Since the difference in that is insignificant this is the number that will be used in the output.
To prevent the issues with the size of the files, the output has been created in the separate file. The example output used for the results has 10000 iterations, however the model is made in a way that custom reports should be easy to generate.

Assignments WS 2020/2021

2020-12-19T22:06:24Z

Doum01: /* Castle Siege */

{{Ambox
| text = <div>
Please, put here your assignments. Do not forget to sign them. You can use <nowiki>~~~~</nowiki> (four tildas) for an automatic signature. Use Show preview in order to check the result before your final sumbition.
</div>
}}

{{Ambox
| text = <div>
Please, strive to formulate your assignment carefully. We expect an adequate effort to formulate the assignment as it is your semestral paper. Do not forget that your main goal is a research paper. It means your simulation model must generate the results that are specific, measurable and verifiable. Think twice how you will develop your model, which entities you will use, draw a model diagram, consider what you will measure. No sooner than when you have a good idea about the model, submit your assignment. And of course, read [[How to deal with the simulation assignment|How to deal with the simulation assignment]].
</div>
}}

{{Ambox
| text = <div>
Topics on gambling, cards, etc. are not welcome.
</div>
}}

{{Ambox
| type = content
| text = <div>
In order to avoid possible confusion, please, check if you have added '''approved''' in bold somewhere in our comment under your submission. If there is no '''approved''', it means the assignment was not approved yet.
</div>
}}

== Spread of 19-COVID ==
'''Author''' : [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 11:32, 11 December 2020 (CET)

'''Simulation''' : I will build a dynamic model of an epidemic his the dynamic issues. I will focus on the spread of Covid-19 in a population and Specially I will try to modelize the effect of a lockdown on the number of infections and deaths.
I will start with a susceptible population, with one infected people that infect 10 people with a certain probability. When you are infected you go the doctors and with a certain probability you go in the recovery population or in the hospital. Then you have a percentage of chance to die in this hospital or to recover. The lockdown will triggered when the ratio of people who where to the doctor and total population exceed a certain percentage. We lockdown and by reducing the number of people you infect in the beginning.

'''Goal of simulation''' : I aim to model the impact of a lockdown on the spread of COVID on a population. It will also show the difference between countries, for example if we lockdown too late. What will be the impact of these decisions.

'''Method''' : I will use VensimPLE for this simulation, as it is a dynamic system, and In my home university I know that I will need to use VensimPLE for a future project and course so it will help me a lot.

::The topic souds good - the question is, on what data sources will you base your simulation (equations) on? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 22:13, 12 December 2020 (CET)
:::I will use the case of Switzerland and use data from this website (https://corona-data.ch). I will try to reproduce the number of cases and the speed of the spread.[[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 13:39, 13 December 2020 (CET)
::: OK. '''Approved''' but do not forget to include in the report for the simulation how you have derived the equations in the simulation from the data. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 22:49, 14 December 2020 (CET)

== NFL Free Agency ==
'''Author''' : [[User:TimWalenczak|TimWalenczak]] ([[User talk:TimWalenczak|talk]]) 14:14, 18 December 2020 (CET)

'''Simulation''' : Free agency is a period during the off-season in the National Football League in the US. During that period, all 32 teams in the league can sign active players that have no contract with any other team, so-called free agents. The process goes as follows: at the beginning of free agency a certain amount of players become free agents due to expiring contracts. All players belong to one out of 8 position groups and want to receive a certain salary. Their starting position on the simulation grid is randomly assigned and they will move over the grid. The overall number of free agents is defined manually while they are randomly distributed over all position groups, as well as their desired salary, which is within a certain manually defined range. Whereas all the teams are looking for players from certain position groups and have a certain amount of money they can spend on them. The teams have fixed positions on the grid and do not move during the simulation. The maximum number of players a team is looking for is defined manually, while the desired position group is defined randomly from the list of position groups and the money to spend is randomly distributed over a certain manually defined salary range. When the simulation starts, the initial demand and supply ratio of a specific position is calculated and affects a player’s value. This means a player’s desired salary rises or decreases due to that demand and supply ratio. During the simulation players move over the grid. Players and teams get in touch when they are located next to each other. When a team’s position group needs matches the player’s position group and a player’s salary fits the money a team offers, the team signs this player and he disappears from the free agency market (turtle dies). If the offer doesn’t fit the need, a player walks on to the next team (turtle takes a random angle turn and walks until it approaches another team). Each team can negotiate only with one player at a time, and vice versa. With every round/tick/signing the demand and supply ratio is recalculated and the desired salary of all players in each position group changes and so does the amount of money teams are able to spend on further players (assuming they just signed a player). Free agency ends after a distinct number of ticks or even earlier if there are no more players left or teams do not have any money left to spend.

'''Goal of simulation''' : is to find out how a player’s value changes during the period of free agency. Especially interesting is, whether players that sign a new contract early in the period received higher payments or the ones signing late in the period. Maybe a certain strategy can be deducted for players depending on their initial demand and supply ratio in order to reach a highly paid contract.

'''Method''' : Since I don’t know VensimPLE yet, I think I will use NetLogo, because I see some similarities between my case and the “Market Structure” case we modelled recently.

: It is a bit unusual, but I can see something in common rather with Market Structure than with Building Escape. Please, just elaborate it into a greater detail and specify, how exactly should your simulation work. I am not sure you have a coherent idea about the solution. But if you work this out, I don't see a problem. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 21:45, 11 December 2020 (CET)

: Yeah you are right I meant the Market Structure, sorry for the confusion! Okay, I updated the description above trying to go more into details. [[User:TimWalenczak|TimWalenczak]] ([[User talk:TimWalenczak|talk]]) 14:14, 18 December 2020 (CET)

:: Ok, '''approved'''. Just bear in mind that as you deal with the random movement, it is a factor that influences the behavior of the model itself. Hence, the way how you implement the movement could be important for the results. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 14:18, 19 December 2020 (CET)

==COVID-19 mass testing==

'''Author''' : [[User:Milan P|Milan P]] ([[User talk:MilanP|talk]]) 20:15, 13 December 2020 (CET)

'''Simulation''' : I will create simulation for mass Covid-19 testing in Slovakian village or small city. In previous rounds of this real mass testing in Slovakia testing itself took place in one day and in one sampling point, this will be the same with this simulation. In this model roles (medical staff, administrative workers, and police officer), will have assigned costs to them in order to include financial aspect into the simulation. Duration of the simulation will be 4 hours. This will serve as base model which will be adjusted in two ways. Firstly, I will create another model in which I will duplicate sampling point so 2 sampling points will be in simulation. Second adjustment to the base model will be that simulation time will be prolongated to 8 hours. After running all 3 models, outputs will be compared and the optimal one will be selected.

'''Goal of simulation''' : In the previous rounds of COVID-19 mass testing in Slovakia certain sampling points had experienced massive overload and were not prepared for the amount of citizens incoming to be tested. Therefore, this simulation would aim to simulate the scenario with data from previous testing rounds and might be helpful if the governments decides to repeat mass testing again.

'''Method''' : Discrete simulation - SIMPROCESS

: The topic looks generaly good, my only concern is that is could be too simple. What steps of the model do you plan? Where do you obtain the real data? [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 02:54, 17 December 2020 (CET)

: I plan to obtain data from this site https://www.svidnik.sk/oznamy/druhe-kolo-plosneho-testovania-vysledky.html - on this website number of patients to be tested and number of points/places that take samples can be found. I know that each point/place has to have: 1 healthcare worker, 2 administrative workers and 1 police officer. Data about hourly rate can be obtained from this site: https://www.platy.sk/platy. Maybe I would like to adjust my original proposal, there would not be 1 point/place taking samples but 9 and I would create another simulation for more or less points/places. Generally I like the idea of splitting patients into two groups and I might include that in my simulation as well. One would consist of people older than 60 and the other one would consist of the rest of population. I will split the population according to demographic structure of Slovakia https://sk.wikipedia.org/wiki/Demografia_Slovenska#Vekov%C3%A1_%C5%A1trukt%C3%BAra. Steps I can think of right now: Patients entering, splitting the patients, administrative work, waiting, taking samples, waiting, receiving results, again splitting patients based on the results (negative ones can go home with certificate and positive ones would require additional steps like receiving information what to do next etc.) and the the simulation would end. [[User:Milan P|Milan P]] ([[User talk:MilanP|talk]]) 00:05, 18 December 2020 (CET)

:: '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 14:36, 19 December 2020 (CET)

==Operation of private dental office==

'''Author''' : [[User:Dufa00|Dufa00]] 10:50, 19 December 2020 (CET)

'''Simulation''' : I would simulate data and conditions of a real private dental office situated in central Bohemia. The Simulation will consist of the aspects influencing the company operations from within and from the outside. Nowadays even the COVID-19 is influencing the revenues and it would be interesting to see how it affects the company.

'''Goal of simulation''' : To help the specific dental office to react on possible dangers and see the hidden opportunities that might be available once the simulation is done. Whatsmore the results should be appliable to any other average private dental office.

'''Method''' : VensimPLE

about nanorobots vs. cancer: An interesting idea. Unfortunately, because this is completely unrealistic, you cannot work with any reasonable data and it means that you are not able to prove the meaningfulness of the results. Please suggest something else. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 03:04, 17 December 2020 (CET)

update: I see, I was thinking that this might happen, is the new proposal better? [[User:Dufa00|Dufa00]] 11:03, 19. December 2020 (CET)

:: If you really have the data for it, it sounds OK. Nevertheless, can you specify the simulation in more detail - what would be the input parameters, variables, what all factors will be incorporated? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 17:27, 19 December 2020 (CET)

==Widening the spread between rich and poor==

'''Author''' : [[User:Sára|Sára]] 21:00, 14 December 2020 (CET)

'''Simulation''' : After 2007-9 world economic crisis a new monetary policy, rather unconventional, was established to escape recessions/depressions. It is called quantitative easing. Although this tool succeeded in reverting the depression and proved helpful during coming years, it is highly discussed what side effects this tool has. A lot of researches suggest that it widens the gap between rich and poor people and make prices of bonds and stocks highly overheated.

'''Goal of simulation''' : Based on macroeconomics theory and semantic experiences from quantitative easing and its influence on other macroeconomic agents like interest rate, inflation, monetary supply etc., i would like to simulate behavior of banks and people during quantitative easing and i would like to work with different values of macroeconomic agents and bank and people behavior in order to try to find out whether widening the spread between rich and poor is an unevitable effect, whether it is sustainable in long term and whether they are possibilities how to erase this side effect.

'''Method''' : NetLogo

: I like the topic (although quantitative easing is just an eufemism for money printing, hence the implications are quite predictable). Before I finally approve, please elaborate it deeper. How exactly you will simulate, what will be the agents, what will they do, etc. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 02:47, 17 December 2020 (CET)

: Agents will be banks and people, until now quantitative easing is been applied always in the situation when interest rates were low for example, it makes people more willing to borrow but banks less willing to lend so idea behind my research is to simulate the process of quantitative easing under different circumstances than it is normally. According to different circumstances i will change people's and banks's behaviors according to macroeconomics theory and empirical knowledge. Is that deep enough? [[User:Sára|Sára]] ([[User talk:Sára|talk]]) 21:01, 18 December 2020 (CET)

:: I am afraid, it is not. What different circumstances would you test? Remind the level of detail of the assignments for our agent-based simulations. It should be quite clear what and how you will simulate. I am still not sure.
:: BTW, quantitative easing is being used while low interest rates, because central banks have no other instrument to use. If the IR were high, it would typically be simpler to decrease it. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:03, 19 December 2020 (CET)

:: Okay so i will simulate wealth increase of the 5% richest households which own 40% of stocks in 20 years compared to estimated world GDP increase if quatitative easing continues in the current way, also i will compare the increase of stocks prices compared to increase of estate prices if this process continues. And i will simulate the situation in which banks would at least 90% of money they get (after central bank buys their bonds and stocks) they will give to people in form of loans and obtain the same numbers as in the first case. Agents will be banks (deciding how much to invest how much to convert into loans and where to invest, based on real numbers), richest households (deciding whether to establish a new company and go on the market with it, whether to release new stocks etc) and central bank (buying stocks and bonds from banks in exchange for money), there will be some probability of recession in which case central bank will release bigger amount of money and will be willing to pay more for the stocks in order to start economy. [[User:Sára|Sára]] ([[User talk:Sára|talk]]) 21:26, 19 December 2020 (CET)

==Rhino population in Africa==

'''Author''' : [[User:Michal F.|Michal F.]] ([[User talk:Michal F.|talk]]) 17:23, 15 December 2020 (CET)

'''Simulation''' : Rhinos once roamed many places in Europe, Asia, and Africa. At the beginning of the 20th century, about 500 000 rhinos roamed Africa and Asia. By 1970, rhino numbers dropped to 70,000, and today, around 27,000 rhinos remain in the wild. Africa's rhino population could face extinction within 10 years, animal welfare experts have warned. South Africa has the largest population of the species in the world however, the species still remains under threat from poaching for its horn and from habitat loss and degradation.
Data: http://www.rhinoresourcecenter.com/pdf_files/119/1192816187.pdf, http://www.scielo.org.za/pdf/koedoe/v59n1/15.pdf, https://rhinos.org/about-rhinos/rhino-species/black-rhino/

'''Goal of simulation''' : Based on the available data, I would like to simulate future development of the rhino population (probably the black ones in Africa, cause its most documented). The population is threatened by poaching and the loss of the natural environment. The shrinking population increases people's efforts to protect it from poachers but increases the demand for rhino horns.

'''Method''' : VensimPLE / Insight Maker

:: Topic is interesting. Can you specify the simulation in more detail - what would be the input parameters, variables, what all factors, that have an influence on rhino population, will be incorporated? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 14:55, 16 December 2020 (CET)

==Garbage collection==

'''Author''' : [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 23:07, 16 December 2020 (CET)

'''Simulation''' : My aim is to simulate garbage collection in a city from the city counciler's point of view. I will focus on 3 main standpoints, specifically Waste production, Waste collection and Waste processing. The amount of produced waste is dependent on the population of the city and the rate of waste production of 1 person per day. Waste collection will be dependent on the amount of garbage trucks and their capacity. The number of employees (drivers and garbage collectors) will also play a role in how many trucks per day can be deployed. The final standpoint, Waste processing, consists of waste incineration and recycling. Incineration and recycling facilities can process different amount of waste per day. The scope of simulation can be either in weeks or months.

'''Goal of simulation''' : Finding balance between the amount of produced waste and its disposal by regulating the number of employees, trucks, their capacity, and the number of facilities which can either incinerate or recycle waste depending on the type of waste. The population of the city and the waste production of 1 person will be also adjustable.

'''Method''' : VensimPLE

::Sounds complex (maybe too complex) - but on what real data will you base the above described simulation on, so that the results would make sense? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 08:14, 18 December 2020 (CET)

Upon further reevaluation I propose to make the model simpler by crossing out the Waste collection standpoint, so there will be Waste production and Waste processing left. On these standpoints I found a valuable source of data: https://www.mzp.cz/cz/odpady_podrubrika . Also, thanks to the available data from 2009 to 2019 I could model the probable evolvement of waste production and waste processing either in Prague or the Czech republic. [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 23:58, 18 December 2020 (CET)

::'''Approved''' and do not forget to include in the report for the simulation how you have derived the equations in the simulation from the data.[[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 13:48, 19 December 2020 (CET)

==Trends of weddings in the future==

'''Author''' : [[User:netp02|netp02]] ([[User talk:netp02|talk]]) 22:30, 17 December 2020 (CET)

'''Simulation''' :

The purpose of this simulation is to show trends and development of weddings in the Czech republic. It will be based on real data available on
Czech Statistical Office. It will predict the expected number of weddings in the future.

'''Goal of simulation''' :

Make a prediction of the trends of weddings for next years based on real data from the Czech Statistical Office from the past and consider all factors which influence the number of weddings in the Czech republic.

'''Method''' : VensimPLE

==IT Team simulation==

'''Simulation:'''

I would like to create a simulation, which is based on a Process Diagram and which would describe the process of solving business requirements within the IT team in which I now work. The simulation would show the entire course of the process from the origin of the request to the passage of all phases To Develop - Code review - Testing - Acceptance - UAT - Done.

'''Goal:'''

By simulation, I would deal with limited resources and limited specialization in the team. The goal would be to find out how many requests and in what rhythm our team would be able to deliver. As for the data basement - It would be based on the situation for the last two months. We solve projects in Jira and so I would be able to calculate how long the request lasted and when it returned to the previous stages + some average morbidity of team members, etc.

'''Method:''' SimProcess simulation

: A bit unusual, but I am curious. '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:06, 19 December 2020 (CET)

==COVID 19 symptoms==

'''Author''' : [[User:Stek15|Stek15]] ([[User talk:Stek15|talk]]) 11:31, 18 December 2020 (CET)

'''Simulation''' :

The model will represent COVID 19. The nodes in the model will represent the common symptoms of COVID 19. According to the World Health Organisation (WHO, 2020) there are 8 common symptoms: fever, dry cough, fatigue, shortness of breath, loss of appetite, confusion, persistent pain or pressure in the chest, high temperature (above 38 °C). The symptoms of said disease are thought to have direct causal relations with one another. If a person develops a symptom of COVID 19 (e.g. shortness of breath), this increases the likelihood of them developing other symptoms (e.g. loss of appetite, fever). Similarly, if one symptom disappears, the others might as well. The greatest danger is present for people vulnerable to the disease. The model will predict whether a person is vulnerable (due to high age) or resistant to the disease and take this in consideration.

'''Goal of simulation''' :

Each tick the probability of a symptom being developed will be calculated. The probability will depend on certain parameters + the total activation of the neighbouring symptoms at the previous tick. These parameters will be the supposed connection between the nodes (slider), influences unrelated to the disease (slider; said age or other respiratory diseases) and the threshold of each symptom.

'''Method''' : NetLogo

: Looks interesting, please, bear in mind, that it is necessary for the model to be based on solid data/parameters to be meaningful. '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:22, 19 December 2020 (CET)

==Investment decision maker==

'''Author''' : [[User:mico00|mico00]] 18:13, 18 December 2020 (CET)

'''Simulation''' : the user has a limited amount of income, and can put in is certain funds. using monte-Carlo simulation, I'd like the user to be able to spread his funds to certain portfolios. these are prone to different oscillations and behavior throughout time. this becomes essentially a game, where the user has the power to work with his money, with the goal to maximize his profits after 10 year period. each transaction is punished by a certain amount of money for time costs.

'''Goal of simulation''' : Main goal is to learn how to handle money in an environment of uncertainty

'''Method''' : google sheets (essentially excel)

:: This is too general. There were some simulations like that in the past, just look for them in previous years. For this kind of task you have to specify what precisely you will simulate, where you will get the data for it (you will derive the probability distributions from), and what new features does your simulation brings compared to the assignments done in the past at this course. The good real data are a must for the MonteCarlo simulation. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 19:36, 18 December 2020 (CET)

==Lawn Mower Roboter==

'''Author''' : [[User:Justin adenau|Justin adenau]] ([[User talk:Justin adenau|talk]]) 19:42, 18 December 2020 (CET)

'''Simulation''' : The Simulation will show the movement and efficiency of a lawn mower robot in different environments (gardens) with different obstacles. It will take attributes like frequency of lawing, wished height of grass, duration of lawing and the power consumption into account. Different algorithms for the movement and the treatment of obstacles will be tried. The limit for the lawing could be a wished height with a defined deviation which is not allowed to be exceeded. The data for the power consumption and duration of lawing can be extracted from variuos providers.

'''Goal of simulation''' : Optimize the lawing frequency and selected lawing height, so that the power consumption is the lowest and the deviation of the wished height of the grass is not exceeded.

'''Method''' : NETLogo

: OK, please, be more specific. What algorithms for the movement will you use? I would also add a goal of testing the individual algorithms and their efficacy. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:50, 19 December 2020 (CET)

: I would try a algortithm which changes the direction randomly when it meets obstacles, algorithms that go systematically around obstacles (assumed that the roboter has a camera to recognize obstacles in short distance) and algortihms that try to go vertically or horizontally by going in lines and go to the next line when hitting an obstacle. The efficiency would be measured by the needed time and power consumption. [[User:Justin adenau|Justin adenau]] ([[User talk:Justin adenau|talk]]) 21:21, 19 December 2020 (CET)

==Retirement Planning and the impact of inflation==

'''Author''' : [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 19:59, 18 December 2020 (CET)

'''Simulation''' : Investing in shares can contribute a great deal to old-age provision. Therefore, I would like to simulate the portfolio decision of a young family. They already have a portfolio of shares and plan to continue investing a fixed amount each year. The decision is to continue to invest in a fixed stock index (SP500) or to invest in a portfolio of 4 CTX stocks from now on.

Data Source: Yahoo Finance

'''Goal of simulation''' : The aim of the simulation is to compare the two portfolios and calculate their probable value in 30 years and what interest income the family can expect in 30 years.

'''Method''' : Excel / Monte Carlo

:: The idea is OK, but the simulation is too simple. Try to extend the simulation proposal so that it is more real. There are repeating financial crises, one can diversify one's portfolio (invest not only in stocks), effect of inflation etc. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 14:25, 19 December 2020 (CET)

I understand. After my research I am not sure if the simulation of stock market crisis is possible with MC. Therefore, I would like to add a third Option of a Portfolio which consists of Bonds and Shares. Additionally I would like to simulate the Inflation rate and check its impact on the different options. [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 22:31, 19 December 2020 (CET)

==Covid 19 - Contract Tracing ==

'''Author''' : [[User:Carolin|Carolin]] ([[User talk:Carolin|talk]]) 21:19, 18 December 2020 (CET)

'''Simulation''' : The Covid 19 global pandemic has an impact on all our lives. Especially the lockdowns in all countries - including the Czech Republic and Germany. The main goal of these lockdowns is to reduce the number of contacts so that contacts of infected people can be traced and chains of infection can be broken. In my simulation, I would like to show the impact of higher distances between people and the prohibition of large public events on the number of infected people.
Contacts with infected people are represented by links, infections are shown by recoloring of the agent. The simulation will show the effects when wearing masks and when not.By this the simulation shows the effect in reducing infections by keeping distance, banning mass events and wearing masks.

'''Data Source''': Data on infection incidence sufficiently available - use only scientific sources

'''Goal of simulation''' : Showing the importance of keeping distance and holding onto rules in a global pandemic, to show the impact everbody has in these times.

'''Method''' : Netlogo

: '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 21:26, 19 December 2020 (CET)

==Evolution of retirement pay in the Czech Republic ==

'''Author''' : [[User:Syca00|Syca00]] ([[User talk:Syca00|talk]]) 22:13, 18 December 2020 (CET)

'''Simulation''' : Nowadays a lot of people say that the Czech Republic will have problems with the retirement pay in the future. So, I decided to create a simulation, which will show a prediction of average retirement pay in the Czech Republic for next 50 years. The average retirement pay depends on more factors and I decided to predict it depending on GDP, state debt, population, mortality and natality.

'''Data Source''': I will use data from mfcr.cz and czso.cz

'''Goal of simulation''' : The aim of the simulation is to predict the evolution of average retirement pay in the Czech Republic and to find out how much money people will get in the future.

'''Method''' : VensimPLE

:: Topic is interesting, but the suggested simulation is too simple for such complex problem. Can you extend the simulation specification so that it matches the reality? (simulation of age groups, gross average wage, etc.) [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 13:56, 19 December 2020 (CET)

== Castle Siege ==

'''Author''' : [[User:Doum01|Doum01]] ([[User talk:Doum01|talk]]) 11:34, 19 December 2020 (CET)

'''Simulation''' : The simulation includes field area and the castle area, attacker agents will have to cross the field area which has an adjustable number of barricades and reach the gate in the castle area then sufficiently damage it to break it. Defender agents will shoot arrows from the castle walls to hinder attackers' progress and protect the gate and subsequently the castle. Attackers and defenders will have multiple types.

'''Data Source''': Number of attackers and defenders and value of their attributes will be adjustable by a common user.

'''Goal of simulation''' : Allowing user to generate custom scenarios and saving their results.

'''Method''' : Netlogo

: sounds like a game. I don't see a point why you should simulate something like that. What should be the output? [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 22:00, 19 December 2020 (CET)

: If I change the setting to ship landing (Day D f.e.) and change the output to find the lowest possible offensive force to have 95% success rate of the siege, will that meet the requirements or shall I change the topic completely ? If this topic won't get accepted anyway my other idea is to simulate a forest ecosystem and optimize at what animal counts should huntsman hunt them to keep the balance in check. [[User:Doum01|Doum01]] ([[User talk:Doum01|talk]]) 22:15, 19 December 2020 (CET)

== Hunting Sharks Simulation ==

'''Author''' : [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 11:56, 19 December 2020 (CET)

'''Simulation''' : I would like to create a simulation of life in the ocean, specifically shark hunts. In the simulation, there will be two types of agents, ''fish'' forming a shoal and ''sharks'' in an adjustable number. The first step will be to implement a swarm / shoal movement algorithm for ''fish'' agents. The ''fish'' will try to avoid the ''sharks'', although they will be slower, but much more mobile in changes of direction. They will also randomly spawn and die. ''Sharks'' will have two modes controlled by hunger. If the ''shark'' is full, it will just roam the ocean. If the ''shark'' is hungry, it switches to hunting mode. This is where the fun begins. The ''shark'' speeds up and starts hunting. When moving in one direction, its speed may increase even more. A ''shark'' in hunting mode will also have a greater vision range than ''fish''.
If there are more ''sharks'', it would be possible to implement reproduction or more types of food for ''shark'' (some can be more attractive for them).

'''Data Source''': Data traceable on the Internet, such as the swimming speed of sharks and common fish.

'''Goal of simulation''' : We do know that sharks are solitary animals, for the most part of live. They typically live and hunt by themselves, joining up with other sharks only in certain circumstances, such as mating. I would like to answer the question of whether it would be beneficial for sharks to form "packs".

'''Method''' : NetLogo

: The topic is good, the goals is meaningless. You will recommend sharks to change their instincts? Please, let me know how and I will approve immediately. Otherwise try to find a better goal. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 22:04, 19 December 2020 (CET)

: Thank you for response. Maybe i can add third specie lets say sea lion or dolphin and new goal will be to create an ecosystem where all species can survive for at least 1000 iterations or create balance. Too many sharks will hunt down sea lions, small numbers of fish will not reproduce fast enough to feed predators. All parameters like fish spawn rate, energy etc. will be adjustable. Or if you have some approvable idea im open for suggestions :) Thank you for answer. [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 22:53, 19 December 2020 (CET)

== Risk Simulation in Banking ==

'''Author''': [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 13:40, 19 December 2020 (CET)

'''Simulation''': Since the Financial Crisis 2007/08 the systematic relevance of banks came into public focus. Excessive risk taking together with an interdependent banking sector led to a dangerous snowball effect which in most countries could only be fixed by massive bail-outs programs.
The exchange of liquidity in between commercial banks on the interbank market, while being beneficial in normal times, offers a treat to the systems stability during black swan events. I therefore plan to simulate the banking market in an arbitrary country and test its resilience to said events with different grades of safety mechanisms (like the Basel III rules for Capital requirements, leverage ratio and Liquidity requirements) in place.

'''Goal of Simulation''': The goal of the Simulation is to access the effectiveness of different measures that were introduced to bolster the financial systems stability. At the same time, I will try to fit the simulation parameters to those of the actual german banking market and try to derive the probabilities of a systematic collapse.

'''Data source''': Deutsche Bundesbank, ECB

'''Method''': Since the non-spatial nature of this topic, I think a MC-Simulation is the best method here.

:: Interesting topic. Can you specify the simulation in more detail - what would be the input parameters, variables, what all factors will be incorporated? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 13:52, 19 December 2020 (CET)

:: The input-parameters would be balance sheet information from the banks like leverage-ratio, liquidity or total assets (possibly with a projection over future years). Variables would be the default probabilities of other banks/private customers. Also I think about including certain "black swan" events like the bursting of housing bubbles or corona. [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 14:17, 19 December 2020 (CET)

:: Include the "black swan" events and it is '''Approved'''. Do not forget to describe in detail in the report for the simulation how you have derived the probability distributions for the random variables and the equations in the simulation from the data. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 17:21, 19 December 2020 (CET)

== Effects of VAT exemption abolition to imported low value consignments ==

'''Author''': [[User:Bara15|bara15]] ([[User talk:bara15|talk]]) 20:00, 19 December 2020 (CET)

'''Simulation''': From 1. 7. 2021 EU Customs Codex has abolished the VAT (Value added tax) exemption limit on imported low-value goods (consignments below 22€). This means, that even for such small consignments, the VAT from the imported goods will have to paid in the state of the consumption. The goods below 150€ will still be exempt from standard customs duties, unless the goods are subject to eg. excise or such. The point of my simulation will be to estimate the novel impacts to the foreign trade (mainly import regimes) volume of the Union.

'''Goal of Simulation''': There are two goals of this simulation. Firstly, to determine the effects of the novel introduction to the tax revenue in the Czech Republic. Secondly, the simulation should be also able to estimate, if this novel will have any impact on the foreign trade in general - can we expect a decrease, or will it not affect the amount of imported goods?

'''Data source''': Intrastat, ČSÚ

'''Method''': Monte Carlo simulation

== Comparing multiple airplane boarding methods ==

'''Author''': [[User:jeda00|jeda00]] ([[User talk:jeda00|talk]]) 21:00, 19 December 2020 (CET)

'''Simulation''': Based on the video about different boarding methods from GCP Gray [https://www.youtube.com/watch?v=oAHbLRjF0vo], I would like to simulate and compare different airplane boarding methods found in the video and possible other sources. The simulation would offer results about how long it takes for passangers board the plane or how often are the passangers stuck in queue, idling.

'''Goal of Simulation''': Goal of the simulation would be to find the most optimal airplane boarding method.

'''Data source''': Publicly available boarding methods and possibly my own boarding methods.

'''Method''': NetLogo

: It sounds really great. I am looking forward to it. '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 22:06, 19 December 2020 (CET)

Assignments WS 2020/2021

2020-12-19T21:15:37Z

Doum01: /* Castle Siege */

{{Ambox
| text = <div>
Please, put here your assignments. Do not forget to sign them. You can use <nowiki>~~~~</nowiki> (four tildas) for an automatic signature. Use Show preview in order to check the result before your final sumbition.
</div>
}}

{{Ambox
| text = <div>
Please, strive to formulate your assignment carefully. We expect an adequate effort to formulate the assignment as it is your semestral paper. Do not forget that your main goal is a research paper. It means your simulation model must generate the results that are specific, measurable and verifiable. Think twice how you will develop your model, which entities you will use, draw a model diagram, consider what you will measure. No sooner than when you have a good idea about the model, submit your assignment. And of course, read [[How to deal with the simulation assignment|How to deal with the simulation assignment]].
</div>
}}

{{Ambox
| text = <div>
Topics on gambling, cards, etc. are not welcome.
</div>
}}

{{Ambox
| type = content
| text = <div>
In order to avoid possible confusion, please, check if you have added '''approved''' in bold somewhere in our comment under your submission. If there is no '''approved''', it means the assignment was not approved yet.
</div>
}}

== Spread of 19-COVID ==
'''Author''' : [[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 11:32, 11 December 2020 (CET)

'''Simulation''' : I will build a dynamic model of an epidemic his the dynamic issues. I will focus on the spread of Covid-19 in a population and Specially I will try to modelize the effect of a lockdown on the number of infections and deaths.
I will start with a susceptible population, with one infected people that infect 10 people with a certain probability. When you are infected you go the doctors and with a certain probability you go in the recovery population or in the hospital. Then you have a percentage of chance to die in this hospital or to recover. The lockdown will triggered when the ratio of people who where to the doctor and total population exceed a certain percentage. We lockdown and by reducing the number of people you infect in the beginning.

'''Goal of simulation''' : I aim to model the impact of a lockdown on the spread of COVID on a population. It will also show the difference between countries, for example if we lockdown too late. What will be the impact of these decisions.

'''Method''' : I will use VensimPLE for this simulation, as it is a dynamic system, and In my home university I know that I will need to use VensimPLE for a future project and course so it will help me a lot.

::The topic souds good - the question is, on what data sources will you base your simulation (equations) on? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 22:13, 12 December 2020 (CET)
:::I will use the case of Switzerland and use data from this website (https://corona-data.ch). I will try to reproduce the number of cases and the speed of the spread.[[User:Toscool|Toscool]] ([[User talk:Toscool|talk]]) 13:39, 13 December 2020 (CET)
::: OK. '''Approved''' but do not forget to include in the report for the simulation how you have derived the equations in the simulation from the data. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 22:49, 14 December 2020 (CET)

== NFL Free Agency ==
'''Author''' : [[User:TimWalenczak|TimWalenczak]] ([[User talk:TimWalenczak|talk]]) 14:14, 18 December 2020 (CET)

'''Simulation''' : Free agency is a period during the off-season in the National Football League in the US. During that period, all 32 teams in the league can sign active players that have no contract with any other team, so-called free agents. The process goes as follows: at the beginning of free agency a certain amount of players become free agents due to expiring contracts. All players belong to one out of 8 position groups and want to receive a certain salary. Their starting position on the simulation grid is randomly assigned and they will move over the grid. The overall number of free agents is defined manually while they are randomly distributed over all position groups, as well as their desired salary, which is within a certain manually defined range. Whereas all the teams are looking for players from certain position groups and have a certain amount of money they can spend on them. The teams have fixed positions on the grid and do not move during the simulation. The maximum number of players a team is looking for is defined manually, while the desired position group is defined randomly from the list of position groups and the money to spend is randomly distributed over a certain manually defined salary range. When the simulation starts, the initial demand and supply ratio of a specific position is calculated and affects a player’s value. This means a player’s desired salary rises or decreases due to that demand and supply ratio. During the simulation players move over the grid. Players and teams get in touch when they are located next to each other. When a team’s position group needs matches the player’s position group and a player’s salary fits the money a team offers, the team signs this player and he disappears from the free agency market (turtle dies). If the offer doesn’t fit the need, a player walks on to the next team (turtle takes a random angle turn and walks until it approaches another team). Each team can negotiate only with one player at a time, and vice versa. With every round/tick/signing the demand and supply ratio is recalculated and the desired salary of all players in each position group changes and so does the amount of money teams are able to spend on further players (assuming they just signed a player). Free agency ends after a distinct number of ticks or even earlier if there are no more players left or teams do not have any money left to spend.

'''Goal of simulation''' : is to find out how a player’s value changes during the period of free agency. Especially interesting is, whether players that sign a new contract early in the period received higher payments or the ones signing late in the period. Maybe a certain strategy can be deducted for players depending on their initial demand and supply ratio in order to reach a highly paid contract.

'''Method''' : Since I don’t know VensimPLE yet, I think I will use NetLogo, because I see some similarities between my case and the “Market Structure” case we modelled recently.

: It is a bit unusual, but I can see something in common rather with Market Structure than with Building Escape. Please, just elaborate it into a greater detail and specify, how exactly should your simulation work. I am not sure you have a coherent idea about the solution. But if you work this out, I don't see a problem. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 21:45, 11 December 2020 (CET)

: Yeah you are right I meant the Market Structure, sorry for the confusion! Okay, I updated the description above trying to go more into details. [[User:TimWalenczak|TimWalenczak]] ([[User talk:TimWalenczak|talk]]) 14:14, 18 December 2020 (CET)

:: Ok, '''approved'''. Just bear in mind that as you deal with the random movement, it is a factor that influences the behavior of the model itself. Hence, the way how you implement the movement could be important for the results. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 14:18, 19 December 2020 (CET)

==COVID-19 mass testing==

'''Author''' : [[User:Milan P|Milan P]] ([[User talk:MilanP|talk]]) 20:15, 13 December 2020 (CET)

'''Simulation''' : I will create simulation for mass Covid-19 testing in Slovakian village or small city. In previous rounds of this real mass testing in Slovakia testing itself took place in one day and in one sampling point, this will be the same with this simulation. In this model roles (medical staff, administrative workers, and police officer), will have assigned costs to them in order to include financial aspect into the simulation. Duration of the simulation will be 4 hours. This will serve as base model which will be adjusted in two ways. Firstly, I will create another model in which I will duplicate sampling point so 2 sampling points will be in simulation. Second adjustment to the base model will be that simulation time will be prolongated to 8 hours. After running all 3 models, outputs will be compared and the optimal one will be selected.

'''Goal of simulation''' : In the previous rounds of COVID-19 mass testing in Slovakia certain sampling points had experienced massive overload and were not prepared for the amount of citizens incoming to be tested. Therefore, this simulation would aim to simulate the scenario with data from previous testing rounds and might be helpful if the governments decides to repeat mass testing again.

'''Method''' : Discrete simulation - SIMPROCESS

: The topic looks generaly good, my only concern is that is could be too simple. What steps of the model do you plan? Where do you obtain the real data? [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 02:54, 17 December 2020 (CET)

: I plan to obtain data from this site https://www.svidnik.sk/oznamy/druhe-kolo-plosneho-testovania-vysledky.html - on this website number of patients to be tested and number of points/places that take samples can be found. I know that each point/place has to have: 1 healthcare worker, 2 administrative workers and 1 police officer. Data about hourly rate can be obtained from this site: https://www.platy.sk/platy. Maybe I would like to adjust my original proposal, there would not be 1 point/place taking samples but 9 and I would create another simulation for more or less points/places. Generally I like the idea of splitting patients into two groups and I might include that in my simulation as well. One would consist of people older than 60 and the other one would consist of the rest of population. I will split the population according to demographic structure of Slovakia https://sk.wikipedia.org/wiki/Demografia_Slovenska#Vekov%C3%A1_%C5%A1trukt%C3%BAra. Steps I can think of right now: Patients entering, splitting the patients, administrative work, waiting, taking samples, waiting, receiving results, again splitting patients based on the results (negative ones can go home with certificate and positive ones would require additional steps like receiving information what to do next etc.) and the the simulation would end. [[User:Milan P|Milan P]] ([[User talk:MilanP|talk]]) 00:05, 18 December 2020 (CET)

:: '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 14:36, 19 December 2020 (CET)

==Operation of private dental office==

'''Author''' : [[User:Dufa00|Dufa00]] 10:50, 19 December 2020 (CET)

'''Simulation''' : I would simulate data and conditions of a real private dental office situated in central Bohemia. The Simulation will consist of the aspects influencing the company operations from within and from the outside. Nowadays even the COVID-19 is influencing the revenues and it would be interesting to see how it affects the company.

'''Goal of simulation''' : To help the specific dental office to react on possible dangers and see the hidden opportunities that might be available once the simulation is done. Whatsmore the results should be appliable to any other average private dental office.

'''Method''' : VensimPLE

about nanorobots vs. cancer: An interesting idea. Unfortunately, because this is completely unrealistic, you cannot work with any reasonable data and it means that you are not able to prove the meaningfulness of the results. Please suggest something else. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 03:04, 17 December 2020 (CET)

update: I see, I was thinking that this might happen, is the new proposal better? [[User:Dufa00|Dufa00]] 11:03, 19. December 2020 (CET)

:: If you really have the data for it, it sounds OK. Nevertheless, can you specify the simulation in more detail - what would be the input parameters, variables, what all factors will be incorporated? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 17:27, 19 December 2020 (CET)

==Widening the spread between rich and poor==

'''Author''' : [[User:Sára|Sára]] 21:00, 14 December 2020 (CET)

'''Simulation''' : After 2007-9 world economic crisis a new monetary policy, rather unconventional, was established to escape recessions/depressions. It is called quantitative easing. Although this tool succeeded in reverting the depression and proved helpful during coming years, it is highly discussed what side effects this tool has. A lot of researches suggest that it widens the gap between rich and poor people and make prices of bonds and stocks highly overheated.

'''Goal of simulation''' : Based on macroeconomics theory and semantic experiences from quantitative easing and its influence on other macroeconomic agents like interest rate, inflation, monetary supply etc., i would like to simulate behavior of banks and people during quantitative easing and i would like to work with different values of macroeconomic agents and bank and people behavior in order to try to find out whether widening the spread between rich and poor is an unevitable effect, whether it is sustainable in long term and whether they are possibilities how to erase this side effect.

'''Method''' : NetLogo

: I like the topic (although quantitative easing is just an eufemism for money printing, hence the implications are quite predictable). Before I finally approve, please elaborate it deeper. How exactly you will simulate, what will be the agents, what will they do, etc. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 02:47, 17 December 2020 (CET)

: Agents will be banks and people, until now quantitative easing is been applied always in the situation when interest rates were low for example, it makes people more willing to borrow but banks less willing to lend so idea behind my research is to simulate the process of quantitative easing under different circumstances than it is normally. According to different circumstances i will change people's and banks's behaviors according to macroeconomics theory and empirical knowledge. Is that deep enough? [[User:Sára|Sára]] ([[User talk:Sára|talk]]) 21:01, 18 December 2020 (CET)

:: I am afraid, it is not. What different circumstances would you test? Remind the level of detail of the assignments for our agent-based simulations. It should be quite clear what and how you will simulate. I am still not sure.
:: BTW, quantitative easing is being used while low interest rates, because central banks have no other instrument to use. If the IR were high, it would typically be simpler to decrease it. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:03, 19 December 2020 (CET)

:: Okay so i will simulate wealth increase of the 5% richest households which own 40% of stocks in 20 years compared to estimated world GDP increase if quatitative easing continues in the current way, also i will compare the increase of stocks prices compared to increase of estate prices if this process continues. And i will simulate the situation in which banks would at least 90% of money they get (after central bank buys their bonds and stocks) they will give to people in form of loans and obtain the same numbers as in the first case. Agents will be banks (deciding how much to invest how much to convert into loans and where to invest, based on real numbers), richest households (deciding whether to establish a new company and go on the market with it, whether to release new stocks etc) and central bank (buying stocks and bonds from banks in exchange for money), there will be some probability of recession in which case central bank will release bigger amount of money and will be willing to pay more for the stocks in order to start economy. [[User:Sára|Sára]] ([[User talk:Sára|talk]]) 21:26, 19 December 2020 (CET)

==Rhino population in Africa==

'''Author''' : [[User:Michal F.|Michal F.]] ([[User talk:Michal F.|talk]]) 17:23, 15 December 2020 (CET)

'''Simulation''' : Rhinos once roamed many places in Europe, Asia, and Africa. At the beginning of the 20th century, about 500 000 rhinos roamed Africa and Asia. By 1970, rhino numbers dropped to 70,000, and today, around 27,000 rhinos remain in the wild. Africa's rhino population could face extinction within 10 years, animal welfare experts have warned. South Africa has the largest population of the species in the world however, the species still remains under threat from poaching for its horn and from habitat loss and degradation.
Data: http://www.rhinoresourcecenter.com/pdf_files/119/1192816187.pdf, http://www.scielo.org.za/pdf/koedoe/v59n1/15.pdf, https://rhinos.org/about-rhinos/rhino-species/black-rhino/

'''Goal of simulation''' : Based on the available data, I would like to simulate future development of the rhino population (probably the black ones in Africa, cause its most documented). The population is threatened by poaching and the loss of the natural environment. The shrinking population increases people's efforts to protect it from poachers but increases the demand for rhino horns.

'''Method''' : VensimPLE / Insight Maker

:: Topic is interesting. Can you specify the simulation in more detail - what would be the input parameters, variables, what all factors, that have an influence on rhino population, will be incorporated? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 14:55, 16 December 2020 (CET)

==Garbage collection==

'''Author''' : [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 23:07, 16 December 2020 (CET)

'''Simulation''' : My aim is to simulate garbage collection in a city from the city counciler's point of view. I will focus on 3 main standpoints, specifically Waste production, Waste collection and Waste processing. The amount of produced waste is dependent on the population of the city and the rate of waste production of 1 person per day. Waste collection will be dependent on the amount of garbage trucks and their capacity. The number of employees (drivers and garbage collectors) will also play a role in how many trucks per day can be deployed. The final standpoint, Waste processing, consists of waste incineration and recycling. Incineration and recycling facilities can process different amount of waste per day. The scope of simulation can be either in weeks or months.

'''Goal of simulation''' : Finding balance between the amount of produced waste and its disposal by regulating the number of employees, trucks, their capacity, and the number of facilities which can either incinerate or recycle waste depending on the type of waste. The population of the city and the waste production of 1 person will be also adjustable.

'''Method''' : VensimPLE

::Sounds complex (maybe too complex) - but on what real data will you base the above described simulation on, so that the results would make sense? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 08:14, 18 December 2020 (CET)

Upon further reevaluation I propose to make the model simpler by crossing out the Waste collection standpoint, so there will be Waste production and Waste processing left. On these standpoints I found a valuable source of data: https://www.mzp.cz/cz/odpady_podrubrika . Also, thanks to the available data from 2009 to 2019 I could model the probable evolvement of waste production and waste processing either in Prague or the Czech republic. [[User:Dobj02|Dobj02]] ([[User talk:Dobj02|talk]]) 23:58, 18 December 2020 (CET)

::'''Approved''' and do not forget to include in the report for the simulation how you have derived the equations in the simulation from the data.[[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 13:48, 19 December 2020 (CET)

==Financial budget of football club==

'''Author''' : [[User:netp02|netp02]] ([[User talk:netp02|talk]]) 22:30, 17 December 2020 (CET)

'''Simulation''' :

The purpose of this simulation is to simulate important financial aspects of the football club. How much sponsors, players, transfers of players, money for tickets, number of sold club´s souvenir influence the budget of the clubs.

'''Goal of simulation''' :

Finding a balance between incomes and expenses, so the football club could work without any money donation from owner of the club.

'''Method''' : VensimPLE

:: On what real data will you base the above described simulation on, so that the results of your simulation would make sense? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 08:16, 18 December 2020 (CET)

:: I would take data from football club Slavia Prague, which announce almost all data for publicity. What it be possible to do it like that?
[[User:netp02|netp02]] ([[User talk:netp02|talk]]) 12:01, 18 December 2020 (CET)
::From the public data you cannot get know-how how a football club works on inside and therefore you cannot build complete system model with its feedback loops. If you would map only the financial flows, the system dynamics is no good for that and the result would be some kind of calcualtion, but not a simulation. I would suggest to try something else.[[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 19:27, 18 December 2020 (CET)

==IT Team simulation==

'''Simulation:'''

I would like to create a simulation, which is based on a Process Diagram and which would describe the process of solving business requirements within the IT team in which I now work. The simulation would show the entire course of the process from the origin of the request to the passage of all phases To Develop - Code review - Testing - Acceptance - UAT - Done.

'''Goal:'''

By simulation, I would deal with limited resources and limited specialization in the team. The goal would be to find out how many requests and in what rhythm our team would be able to deliver. As for the data basement - It would be based on the situation for the last two months. We solve projects in Jira and so I would be able to calculate how long the request lasted and when it returned to the previous stages + some average morbidity of team members, etc.

'''Method:''' SimProcess simulation

: A bit unusual, but I am curious. '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:06, 19 December 2020 (CET)

==COVID 19 symptoms==

'''Author''' : [[User:Stek15|Stek15]] ([[User talk:Stek15|talk]]) 11:31, 18 December 2020 (CET)

'''Simulation''' :

The model will represent COVID 19. The nodes in the model will represent the common symptoms of COVID 19. According to the World Health Organisation (WHO, 2020) there are 8 common symptoms: fever, dry cough, fatigue, shortness of breath, loss of appetite, confusion, persistent pain or pressure in the chest, high temperature (above 38 °C). The symptoms of said disease are thought to have direct causal relations with one another. If a person develops a symptom of COVID 19 (e.g. shortness of breath), this increases the likelihood of them developing other symptoms (e.g. loss of appetite, fever). Similarly, if one symptom disappears, the others might as well. The greatest danger is present for people vulnerable to the disease. The model will predict whether a person is vulnerable (due to high age) or resistant to the disease and take this in consideration.

'''Goal of simulation''' :

Each tick the probability of a symptom being developed will be calculated. The probability will depend on certain parameters + the total activation of the neighbouring symptoms at the previous tick. These parameters will be the supposed connection between the nodes (slider), influences unrelated to the disease (slider; said age or other respiratory diseases) and the threshold of each symptom.

'''Method''' : NetLogo

: Looks interesting, please, bear in mind, that it is necessary for the model to be based on solid data/parameters to be meaningful. '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:22, 19 December 2020 (CET)

==Investment decision maker==

'''Author''' : [[User:mico00|mico00]] 18:13, 18 December 2020 (CET)

'''Simulation''' : the user has a limited amount of income, and can put in is certain funds. using monte-Carlo simulation, I'd like the user to be able to spread his funds to certain portfolios. these are prone to different oscillations and behavior throughout time. this becomes essentially a game, where the user has the power to work with his money, with the goal to maximize his profits after 10 year period. each transaction is punished by a certain amount of money for time costs.

'''Goal of simulation''' : Main goal is to learn how to handle money in an environment of uncertainty

'''Method''' : google sheets (essentially excel)

:: This is too general. There were some simulations like that in the past, just look for them in previous years. For this kind of task you have to specify what precisely you will simulate, where you will get the data for it (you will derive the probability distributions from), and what new features does your simulation brings compared to the assignments done in the past at this course. The good real data are a must for the MonteCarlo simulation. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 19:36, 18 December 2020 (CET)

==Lawn Mower Roboter==

'''Author''' : [[User:Justin adenau|Justin adenau]] ([[User talk:Justin adenau|talk]]) 19:42, 18 December 2020 (CET)

'''Simulation''' : The Simulation will show the movement and efficiency of a lawn mower robot in different environments (gardens) with different obstacles. It will take attributes like frequency of lawing, wished height of grass, duration of lawing and the power consumption into account. Different algorithms for the movement and the treatment of obstacles will be tried. The limit for the lawing could be a wished height with a defined deviation which is not allowed to be exceeded. The data for the power consumption and duration of lawing can be extracted from variuos providers.

'''Goal of simulation''' : Optimize the lawing frequency and selected lawing height, so that the power consumption is the lowest and the deviation of the wished height of the grass is not exceeded.

'''Method''' : NETLogo

: OK, please, be more specific. What algorithms for the movement will you use? I would also add a goal of testing the individual algorithms and their efficacy. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 20:50, 19 December 2020 (CET)

: I would try a algortithm which changes the direction randomly when it meets obstacles, algorithms that go systematically around obstacles (assumed that the roboter has a camera to recognize obstacles in short distance) and algortihms that try to go vertically or horizontally by going in lines and go to the next line when hitting an obstacle. The efficiency would be measured by the needed time and power consumption. [[User:Justin adenau|Justin adenau]] ([[User talk:Justin adenau|talk]]) 21:21, 19 December 2020 (CET)

==Retirement Planning==

'''Author''' : [[User:Louis|Louis]] ([[User talk:Louis|talk]]) 19:59, 18 December 2020 (CET)

'''Simulation''' : Investing in shares can contribute a great deal to old-age provision. Therefore, I would like to simulate the portfolio decision of a young family. They already have a portfolio of shares and plan to continue investing a fixed amount each year. The decision is to continue to invest in a fixed stock index (SP500) or to invest in a portfolio of 4 CTX stocks from now on.

Data Source: Yahoo Finance

'''Goal of simulation''' : The aim of the simulation is to compare the two portfolios and calculate their probable value in 30 years and what interest income the family can expect in 30 years.

'''Method''' : Excel / Monte Carlo

:: The idea is OK, but the simulation is too simple. Try to extend the simulation proposal so that it is more real. There are repeating financial crises, one can diversify one's portfolio (invest not only in stocks), effect of inflation etc. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 14:25, 19 December 2020 (CET)

==Covid 19 - Contract Tracing ==

'''Author''' : [[User:Carolin|Carolin]] ([[User talk:Carolin|talk]]) 21:19, 18 December 2020 (CET)

'''Simulation''' : The Covid 19 global pandemic has an impact on all our lives. Especially the lockdowns in all countries - including the Czech Republic and Germany. The main goal of these lockdowns is to reduce the number of contacts so that contacts of infected people can be traced and chains of infection can be broken. In my simulation, I would like to show the impact of higher distances between people and the prohibition of large public events on the number of infected people.
Contacts with infected people are represented by links, infections are shown by recoloring of the agent. The simulation will show the effects when wearing masks and when not.By this the simulation shows the effect in reducing infections by keeping distance, banning mass events and wearing masks.

'''Data Source''': Data on infection incidence sufficiently available - use only scientific sources

'''Goal of simulation''' : Showing the importance of keeping distance and holding onto rules in a global pandemic, to show the impact everbody has in these times.

'''Method''' : Netlogo

: '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 21:26, 19 December 2020 (CET)

==Evolution of retirement pay in the Czech Republic ==

'''Author''' : [[User:Syca00|Syca00]] ([[User talk:Syca00|talk]]) 22:13, 18 December 2020 (CET)

'''Simulation''' : Nowadays a lot of people say that the Czech Republic will have problems with the retirement pay in the future. So, I decided to create a simulation, which will show a prediction of average retirement pay in the Czech Republic for next 50 years. The average retirement pay depends on more factors and I decided to predict it depending on GDP, state debt, population, mortality and natality.

'''Data Source''': I will use data from mfcr.cz and czso.cz

'''Goal of simulation''' : The aim of the simulation is to predict the evolution of average retirement pay in the Czech Republic and to find out how much money people will get in the future.

'''Method''' : VensimPLE

:: Topic is interesting, but the suggested simulation is too simple for such complex problem. Can you extend the simulation specification so that it matches the reality? (simulation of age groups, gross average wage, etc.) [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 13:56, 19 December 2020 (CET)

== Castle Siege ==

'''Author''' : [[User:Doum01|Doum01]] ([[User talk:Doum01|talk]]) 11:34, 19 December 2020 (CET)

'''Simulation''' : The simulation includes field area and the castle area, attacker agents will have to cross the field area which has an adjustable number of barricades and reach the gate in the castle area then sufficiently damage it to break it. Defender agents will shoot arrows from the castle walls to hinder attackers' progress and protect the gate and subsequently the castle. Attackers and defenders will have multiple types.

'''Data Source''': Number of attackers and defenders and value of their attributes will be adjustable by a common user.

'''Goal of simulation''' : Allowing user to generate custom scenarios and saving their results.

'''Method''' : Netlogo

: sounds like a game. I don't see a point why you should simulate something like that. What should be the output? [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 22:00, 19 December 2020 (CET)

: If I change the setting to ship landing (Day D f.e.) and change the output to find the lowest possible offensive force to have 95% succes rate of the siege, will that meet the requirements or shall I change the topic completely ? [[User:Doum01|Doum01]] ([[User talk:Doum01|talk]]) 22:15, 19 December 2020 (CET)

== Hunting Sharks Simulation ==

'''Author''' : [[User:Frym06|Frym06]] ([[User talk:Frym06|talk]]) 11:56, 19 December 2020 (CET)

'''Simulation''' : I would like to create a simulation of life in the ocean, specifically shark hunts. In the simulation, there will be two types of agents, ''fish'' forming a shoal and ''sharks'' in an adjustable number. The first step will be to implement a swarm / shoal movement algorithm for ''fish'' agents. The ''fish'' will try to avoid the ''sharks'', although they will be slower, but much more mobile in changes of direction. They will also randomly spawn and die. ''Sharks'' will have two modes controlled by hunger. If the ''shark'' is full, it will just roam the ocean. If the ''shark'' is hungry, it switches to hunting mode. This is where the fun begins. The ''shark'' speeds up and starts hunting. When moving in one direction, its speed may increase even more. A ''shark'' in hunting mode will also have a greater vision range than ''fish''.
If there are more ''sharks'', it would be possible to implement reproduction or more types of food for ''shark'' (some can be more attractive for them).

'''Data Source''': Data traceable on the Internet, such as the swimming speed of sharks and common fish.

'''Goal of simulation''' : We do know that sharks are solitary animals, for the most part of live. They typically live and hunt by themselves, joining up with other sharks only in certain circumstances, such as mating. I would like to answer the question of whether it would be beneficial for sharks to form "packs".

'''Method''' : NetLogo

: The topic is good, the goals is meaningless. You will recommend sharks to change their instincts? Please, let me know how and I will approve immediately. Otherwise try to find a better goal. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 22:04, 19 December 2020 (CET)

== Risk Simulation in Banking ==

'''Author''': [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 13:40, 19 December 2020 (CET)

'''Simulation''': Since the Financial Crisis 2007/08 the systematic relevance of banks came into public focus. Excessive risk taking together with an interdependent banking sector led to a dangerous snowball effect which in most countries could only be fixed by massive bail-outs programs.
The exchange of liquidity in between commercial banks on the interbank market, while being beneficial in normal times, offers a treat to the systems stability during black swan events. I therefore plan to simulate the banking market in an arbitrary country and test its resilience to said events with different grades of safety mechanisms (like the Basel III rules for Capital requirements, leverage ratio and Liquidity requirements) in place.

'''Goal of Simulation''': The goal of the Simulation is to access the effectiveness of different measures that were introduced to bolster the financial systems stability. At the same time, I will try to fit the simulation parameters to those of the actual german banking market and try to derive the probabilities of a systematic collapse.

'''Data source''': Deutsche Bundesbank, ECB

'''Method''': Since the non-spatial nature of this topic, I think a MC-Simulation is the best method here.

:: Interesting topic. Can you specify the simulation in more detail - what would be the input parameters, variables, what all factors will be incorporated? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 13:52, 19 December 2020 (CET)

:: The input-parameters would be balance sheet information from the banks like leverage-ratio, liquidity or total assets (possibly with a projection over future years). Variables would be the default probabilities of other banks/private customers. Also I think about including certain "black swan" events like the bursting of housing bubbles or corona. [[User:Simon|Simon]] ([[User talk:Simon|talk]]) 14:17, 19 December 2020 (CET)

:: Include the "black swan" events and it is '''Approved'''. Do not forget to describe in detail in the report for the simulation how you have derived the probability distributions for the random variables and the equations in the simulation from the data. [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 17:21, 19 December 2020 (CET)

== Effects of VAT exemption abolition to imported low value consignments ==

'''Author''': [[User:Bara15|bara15]] ([[User talk:bara15|talk]]) 20:00, 19 December 2020 (CET)

'''Simulation''': From 1. 7. 2021 EU Customs Codex has abolished the VAT (Value added tax) exemption limit on imported low-value goods (consignments below 22€). This means, that even for such small consignments, the VAT from the imported goods will have to paid in the state of the consumption. The goods below 150€ will still be exempt from standard customs duties, unless the goods are subject to eg. excise or such. The point of my simulation will be to estimate the novel impacts to the foreign trade (mainly import regimes) volume of the Union.

'''Goal of Simulation''': There are two goals of this simulation. Firstly, to determine the effects of the novel introduction to the tax revenue in the Czech Republic. Secondly, the simulation should be also able to estimate, if this novel will have any impact on the foreign trade in general - can we expect a decrease, or will it not affect the amount of imported goods?

'''Data source''': Intrastat, ČSÚ

'''Method''': Monte Carlo simulation

== Comparing multiple airplane boarding methods ==

'''Author''': [[User:jeda00|jeda00]] ([[User talk:jeda00|talk]]) 21:00, 19 December 2020 (CET)

'''Simulation''': Based on the video about different boarding methods from GCP Gray [https://www.youtube.com/watch?v=oAHbLRjF0vo], I would like to simulate and compare different airplane boarding methods found in the video and possible other sources. The simulation would offer results about how long it takes for passangers board the plane or how often are the passangers stuck in queue, idling.

'''Goal of Simulation''': Goal of the simulation would be to find the most optimal airplane boarding method.

'''Data source''': Publicly available boarding methods and possibly my own boarding methods.

'''Method''': NetLogo

: It sounds really great. I am looking forward to it. '''Approved'''. [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 22:06, 19 December 2020 (CET)

Assignments WS 2020/2021

2020-12-19T10:35:01Z

Doum01: