Assignments WS 2015/2016

From Simulace.info
Revision as of 21:29, 25 December 2016 by Tomáš (talk | contribs) (Traffic simulation(Jakub Vitvar, xvitj27))
Jump to: navigation, search


Brownian motion Stock simulation (Willem Van de Velde, wilv00)

The simulation will contain the evolution of a portfolio of stocks. Each stock has a certain drift, volatility and initial market price. Over time the value of a stock change dependent on a brownian motion and the time. A person wants to invest a certain amount of money in stocks.

The goal of this simulation is to see how the persons initial investment will evolve over time, what portfolio structure would be ideal for the initial investment?

Method: Monte carlo Simulation; Program: Excel

Wilv00 (talk) 09:52, 13 December 2016 (CET)

Hello, could you be more specific? How many stocks? what will be the options for the simulation user to setup his own portfolio? how will you implement the periodically returning depressions? I would expect that the portfolio should consist from stocks, bonds and deposits and the user would choose between the split of the portfolio among these three financial instruments, rather than choosing particular stocks (as it is done in reality). What will be the source of the data?

Oleg.Svatos (talk) 20:48, 13 December 2016 (CET)

I would use the stock market of my home country (Belgium) so the BEL-20. I would do the simulation with the option of selecting a portfolio that consists indeed of bonds, deposits and stocks taking into account with current market intrest from Belgium and returns on these instruments. I would involve 5 different stocks from the BEL-20 and combine these into a portfolio. So the total simulation would involve a part that is invested in bonds, deposits and a portfolio of 5 stocks (i haven't looked which specific stocks i would use). The goal will than be determing which percentage of his initial investment he would have to invest in bonds, stocks and deposists to maximize his return after a certain period of years (taken into account the different risk measures associated with the different portfolio's).

I wasn't thinking about involving periodically returning depressions. This would mean that after a certain period of time I would have to change the volatility and drift of the stocks and the market intrest right?

Wilv00 (talk) 12:09, 14 December 2016 (CET)

Yes, you have to play with it little bit as the value of the instruments/portfolios changes differently during the depression as you can see for instance in graph here. These business cycles is what makes the portfolio value prediction so hard, so it is necessary to incorporate them in the simulation. Simulation approved.

Oleg.Svatos (talk) 13:43, 14 December 2016 (CET)

Ok, i see. I will incorporate this in the model as well. Thank you.

Wilv00 (talk) 09:34, 15 December 2016 (CET)

Simulation proposal (Jakub Esterka, xestj00)

Simulation (in NETLOGO) will be set up as a FPS computer game (e.g. Counter Strike) with two teams and set number of players in each team (between 1 and 64). Goal of a simulation will be to find out how much more members can other team have for winning rate to be still 50 % in average. Second goal will be to find out how big advantage randomly selected player has to have for his team to dominate (after 10 games). And the third goal is to find the average rate of domination – how many games is needed for one team to start to dominate the other in every game (team members have almost maximum accuracy of 100).

- Each team can have 1 – 10 players

- Each team begins opposite of each other

- Each player moves randomly and have random accuracy (1-20)

- One player from each team can have an advantage (accuracy of 80)

- When two players from opposite teams meet, they shoot at each other with random accuracy

- If one player shoots the other one, he has its shooting accuracy increased

- When all players from one team die, game resets but players keep their accuracy (except for the first goal)


First goal is obligatory and in this research players will not keep their accuracy. If the simulation will work as planned and it will be possible for players to keep their accuracy, other goals are viable as well.

--Xestj00 (talk) 18:42, 9 December 2016 (CET)

--- I am not generally very happy about simulations regarding video games and other entirely artificial problems, because a real benefit of such a simulation is more than questionable. But, ok... Moreover, in this case, could we be even sure that the results of the simulation really fit the game? Or in other words, how could we be sure that the parameters mentioned above and the way how you will simulate it matches how Counter Strike works? Tomáš (talk) 01:11, 15 December 2016 (CET)

Software Developers (xvatj00)

Description

There is a small company for software development. The company takes small orders (around 3 months long) for software. The software development process consists of 3 parts - analysis, coding and testing. For each project, there is a set number of analysts, programmers and testers, who work on the given project (and no other) until it’s finished. There are 2 main types of orders - well defined and very generally defined. For well defined orders (30% of all orders), the average time of the process of analysis is shorter than in case of the generally defined order, and requires less analysts. The company knows the average times of all parts of the development as well as the average salaries of the three positions. Each part of the development also has its probability of delay and average delay (in case delay occurs). There are different scenarios in cases of delays. Alongside specific positions, there are also “ultimate” workers, who work on the project from its beginning until the end - meaning they do all 3 parts of the development themselves. These workers have higher salaries (all are seniors), but take less time to finish the work. Method: Simprocess

Goal of the Simulation

The goal is to see, if it’s more convenient to have specialized workers, or to hire “ultimate” workers instead, and to have an overview of the projects’ employee expenses. To reach the goal, the plan is to create 1 simulation which takes into consideration only the process with the positions divided, and a second simulation which shows the process of “ultimate” workers. In both cases, the process will be taking all the incoming orders - there will be only this one type of process.

Jana Vataščinová, Xvatj00 (talk) 20:29, 10 December 2016 (CET)

- Hi, I am missing the tool you want to use. Moreover, I hesitate a bit, whether such a "soft" process like software development should be analyzed this way. Tomáš (talk) 01:15, 15 December 2016 (CET)

- Good evening, I would like to use Simprocess. I was thinking that the information to the process was quite stable, so I chose the topic, because it sounds interesting to me. Xvatj00 (talk) 01:49, 15 December 2016 (CET)

I am sorry, but I don't see an added value in this topic. Software development is really a "soft" field, every project is different, every developer is different, they are specialized for particular things, etc. etc. And, the most important for me: I don't see a chance that anyone would - even theoretically - use your simulation to help their business. Don't take me wrong, I don't expect that all the simulations we do here will be used in a real company for a real purpose, however, in some cases, it would be at least theoretically possible after some elaboration (like the grocery store example). In the case of software development simulation, as you suggested, I really cannot imagine that anyone would ever use it for a real business. Please, either try to find a particular, well structured and "hard" part of the problem and reformulate or try something completely different. Tomáš (talk) 13:29, 18 December 2016 (CET)
Good evening, what about I simulated the process as a whole, without dividing the employees, taking into account average times for each part of the development, chances of delays and their scenarios, etc. Then I could find how many orders I would be able to process at maximum with the current number of employees, and how costly one order is on average? I will try to think of another simulation as well, would it maybe be possible to get some tips for the topic? Xvatj00 (talk) 22:07, 18 December 2016 (CET)
Please, try something else asap. Software development is really very hard to manage this way. Projects are very different even if they could seem relatively predictable (like e.g. a development of various plugins). I am afraid you would have a weak fit with reality. Tomáš (talk) 00:30, 22 December 2016 (CET)

Police System (Kateryna Ushanova, xushk00)

This simulation will show police system and how this system will impact the crime rate of the city. Criminals will try to destroy the city's infrostructre and to make crime against civilians. In this simulation I will use three strategies that maybe used to fight crime:

1) applying medium-sized police force

2) a strategy in which a small police force attempt to recruit civilians into the police force

3) the police react on the crime in a brief period of time


The most important factor that influence skill of the police to fight the crime is work experience. The amount of experience is propotional to the cost of fighting crime.

The criminals will be walking randomly around the visual area. During their walking there is a random chance that they will commit a crime. When they do the crime, the civilians around them become victims, and the area they occupy becomes a bad zone. Bad zone make people more likely to turn to crime. Two main factors influence whether or not civilians will turn to crime:

1)the level of the city (user controls)

2)the simulated factors (the result of the police work; the state of the zone; the effects of random events that the user controls)


The goal of this simulation is:

- to see how changing the work experience level effects each strategy.

- to test out different strategies and factors to see how this effects the crime rate.

- to observe possible outcomes by changing techniques and factors in a different ways.

Method - NetLogo

Kateryna Ushanova, Xushk00 (talk) 13:35, 11 December 2016 (CET)

- The idea is not bad, but where are your key parameters for the simulation? How you get them (I mean the exact figures). We always prefer obtaining and working with real numbers. Sometimes, it is not possible, so it is acceptable to use secondary resources, etc. but it should be always clear, why did you use the parameters you used and why, on the other hand Tomáš (talk) 01:23, 15 December 2016 (CET)

Thank you for your comment. My key parametrs are: number of criminals; amount of poverty in the city (which is directly proportional to the willingness of the civilians to become criminals; level of experience of officers. I will use statistical data from my city.

Xushk00 (talk) 19:38, 15 December 2016 (CET) Kateryna Ushanova

You strive to work with very "soft" characteristics like the experience of police officers, willingness to become a criminal... How would you simulate it? How would you measure it? What source of data would you use? I am afraid it would be very risky for you, because you could have trouble to prove that the model is meaningful. I suggest to find something else. Tomáš (talk) 13:33, 18 December 2016 (CET)

Optimal schedule of Moscow metro (NEW ASSIGNMENT)

The simulation will show situation in Moscow metro in different parts of the day (peak hour and off-peak hour). In this simulation i will find optimal time of train´s delay in concrete line in Moscow metro. The input numbers will be taken from resource with real data.

Key parametrs:

- periodicity of people travalling by metro

- delay time of each train

- number of passengers in the station (it depends on part of the day (e.g. peak hour)

- capacity of train

The main goal is to figure out in which situations traffic jams occurs and find the optimal ratio between number of travellers and time of delay.

Method: NetLogo

Xushk00 (talk) 18:43, 18 December 2016 (CET) xushk00, Kateryna Ushanova

Approved Tomáš (talk) 00:31, 22 December 2016 (CET)

(Poison and Insect Population, Kellianne Stefl)

NetLogo simulation of the steady decline in insect population in Germany from 1989 to now, also future predictions. Many other factors contribute but I will use pesticides in my visual representation.

Over the last 25 years, etymologists have set up tents all over German woodlands and meadows to observe natures insects. The average biomass of insects sampled in a 6-month period (May to October) has steadily decreased from 1.6 kilograms (3.5 pounds) (56 ounces) per trap in 1989 to just 300 grams (10.6 ounces) in 2014. 18.9% of the average biomass 25 years ago. This simulation will illuminate the devastating effects of damage to the bottom of the food chain.

-adjustable poison release

Goal: Predict hypothetical decline of insect population: how many years until there are none left.

Kellistefl (talk) 12:47, 12 December 2016 (CET)

- Ok, we talk about entomologysts, not etymologysts, don't we? The topic is generally good, I am just not sure how exactly it will look like. Particularly, I am a bit afraid that the simulation would be pretty simple (e.g. I can see just one parameter now). Please, could you elaborate a bit? Tomáš (talk) 01:43, 15 December 2016 (CET)

--Yes, sorry for the bad wording. Some things that make the simulation more complicated will be: insect reproduction, each will slowly lose energy as they are poisoned by pesticides and some portion will not be able to reproduce. Sorting out how many ounces of insects each pesticide dispenser (airplane) can kill for a given time will be calculated. It will have to be an amount that decreases population even while a certain percentage of insects are still reproducing. I will attempt to use real life data as much as I can to make the simulation as accurate as possible. Kellistefl (talk) 13:32, 16 December 2016 (CET)

As we discussed, the simulation is rather simple. But it is acceptable if you focus on data, its correct representation, scaling, and overall fit with reality. I prefer a simpler model with perfect output that someone could use for a real problem over a complex model that would be completely useless. But, beware: it is a complicated and risky way. Approved. Tomáš (talk) 13:39, 18 December 2016 (CET)

Phyllotaxis (Nada Bednarova, bedn00)

Phyllotaxis is the process of arrangement of leaves on a plant stem. While growing, leaves on a plant/flower/cone usually follow Fibonacci sequences or use golden ratio to form the most convenient angles. There are several theories on where the arrangement comes from.

Simulation in NetLogo would simulate the arrangement of leaves based on different parameters configuration (number and size of the leaves, starting angle etc.). Furthermore chemical or physiological theories about morphogenesis / meristem growth can be simulated (for example Turing's diffusion equations).

--Bedn00 (talk) 00:07, 15 December 2016 (CET)

This isn't an agent-based model, however the idea looks interesting, so approved Tomáš (talk) 01:45, 15 December 2016 (CET)

The influence of exchange rates and interest rates on importers/exporter (Jakub Vitvar, xvitj27)

My goal is to create multi agent simulation in Netlogo. The simulation will contain sets of producers(agents) domestic and foreign. These producers will focus on importing or exporting based on current economical state and their choice will have impact on others.

The goal of the simulation is to find out whether the market is able to balance itself. Other goal could be to find out how will the market react if we set some variables manualy.

Xvitj27 (talk) 16:02, 15 December 2016 (CET)

Maybe the system dynamics model could be more suitable, since this model would be too complicated for NetLogo. Xvitj27 (talk) 10:39, 17 December 2016 (CET)

Please elaborate the simulation topic further, currently it is too general. Importers and exporters of what kind of goods? On basis of which theory (there are many that are trying to find the equilibrium point) ? What kind of market? Who will be the participants? What is the "economical state exactly" and what effect it should have on the importers and exporters? System dynamics may be applicable, but the further detail for this decision is necessary. Oleg.Svatos (talk) 16:06, 17 December 2016 (CET)

I would follow the process exactly as it is taught here on VSE. Meaning the influences on the enviroment would be monetary and fiscal state interventions on the market. The number of the exporters importers would be variable. The type of participants and type of goods should be irrelevant. The importers/exporters would react on changes in money supply/demand change after the intervention. The goal would be to show the next equilibrium and how long does it take to find. I am still ready to change this subject completely if it seems not appropriate for this assignment. Xvitj27 (talk) 16:54, 17 December 2016 (CET)

So far the description, you have given, is very abstract - the causalities seem unclear to me. That's why I have asked whether you have it based on some firm theory and that you are confident you can create a simulation on basis of it. If you do, go for it, if not, I would rather pick some less abstract topic as it might be easier to develop the simulation. Oleg.Svatos (talk) 10:02, 18 December 2016 (CET)

Would it be ok, if I tried doing this simulation based on macroeconomic textbook from VSE and If I find out that it is too complicated I will do some other simulation? I am already thinking about trafic simulation in netlogo with simple trafic network and agents using A* algorithm to find way around some accidents and trafic closures. Xvitj27 (talk) 20:18, 18 December 2016 (CET)

The rules are in this case strict (and there is not enough time for this "what if approach" anyway). So you have to choose the simulation topic now. Oleg.Svatos (talk) 08:31, 19 December 2016 (CET)

And the trafic simulation would be okay? Xvitj27 (talk) 11:01, 19 December 2016 (CET) Should I create new simulation assignment here? Xvitj27 (talk) 11:44, 19 December 2016 (CET)

Yes, you have to create new topic here which will contain all the required information. Oleg.Svatos (talk) 14:36, 19 December 2016 (CET)

Corporate Tax Rate's effect on capital influx (André Seyffarth Årnes, aara00)

I want to create a simulation in NetLogo that shows how different corporate tax rates in Europe affects the capital influx of countries. The initial simulation should start with the tax rates that exists in Europe today, and the simulation should show what happens when they change. In this sense the simulation will only include two different actors, namely nations and companies.

I should also be able to relate this to game theory.

Aara00 (talk) 13:28, 16 December 2016 (CET)

It is a very interesting problem but could be very complicated. Please, elaborate on how exactly would you solve it. Tomáš (talk) 13:41, 18 December 2016 (CET)

Post office check-out optimalization (Josef Zeman - xzemj34)

Software: Simproces

Model: This simulation is going to be focused on optimalization of open check-outs (counters) at the post office. Given the fact these simulations should be relatable to the real world, I chose ideally the post office of Česká pošta in Jindřišská street (if this one doesnt work out for me, i'll look for another one) where I'd like to get the real numerous data about the visitors from the "call-out infosystem". This system distributes visitors accordingly to the capacity of the office as well as to selected issue visitiors want to solve (send package, pickup package, czech point, pensions etc.) These data should vary a lot during month since there are certain periods every month when many people have to visit post office (e.g.pensions and social benefits). Every counter can serve for at least one of the issue mentioned above, most likely for more than one, some may serve to all of them.

Goal of the simulation: Main point is to ideally and optimally assign human resources to counters accordingly given the daytime and the period of month.

Good idea, but it is invetable to obtain data from a real post office. Doesn't necessarily need to be from the main post office. Approved. Tomáš (talk) 13:43, 18 December 2016 (CET)

School is on Fire (Zbyněk Zelinka, xzelz06)

A simulation in NetLogo describing an evacuation of the "Stará budova" at VŠE. The layout and proportions of the building will be taken from the "VŠE Maps" project. I consider including just the first floor and adding an inflow of people from higher floors. I also intend to base it on a real schedule of the classrooms and compare the outcomes from multiple time periods. There will also be fire patches in the model, limiting the space people can walk in.

The evacuated people will have different speeds and reaction times, also some of them won't know which way to go. I want to measure the time it takes to get everybody to the safe zone outside the building and find the weak spots where the big crowds get stuck.

xzelz06 (talk) 20:54, 16 December 2016 (CET)

OK, sounds good. But, don't forget to consider real speeds of people, some intelligent movement, a speed of spreading fire, obstacles, etc. etc. Could be quite complex, but very interesting. Approved. Tomáš (talk) 13:46, 18 December 2016 (CET)

Spreading of a fatal epidemic in a simple society (Filip Vacula, xvacf01)

Software: Vensim

Description: A system dynamics simulation, modelling the behaviour of a simplified society on the verge of an epidemic. The model is much simplified as opposed to the reality, which would be almost impossible to model. However, my main objective is to keep the main and most important factors in play, to be able to observe their behaviour.

Goal: The main goal is to observe the effect of distribution of society to 3 categories based on their yearly income, and how much can the spreading of the epidemic be influenced by the health care system in place (government-funded vs. government-unfunded). However, with a highly parametric model, the effect of many more variables can be easily observed. A side goal is to observe the most critical interactions inside the system, which could lead to a faster treatment of the population in case of such epidemic.

Model environment: The epidemic has a chance of transferring to another person by him being in the same room as an infected person for at least 30 minutes. Such encounters between people daily would be from a discrete interval. The chance is further lowered by a vaccination, which becomes available on the 30th day after the start of the epidemic, however has only a small efficiency, and is also quite costly. A highly effective treatment becomes available 180 days after the epidemic has started, but is again very costly. The production of both the vaccines and treatment takes roughly 30 days to fully satisfy the demand of the population (linearly).

Xvacf01 (talk) 21:59, 16 December 2016 (CET)

OK, approved, just be aware of the the fact that the Vensim cannot do spacial simulations (like people moving in some space).

Oleg.Svatos (talk) 15:56, 17 December 2016 (CET)

The strength of tsunami (Hana Nápravníková, xnaph00)

NetLogo simulation describes the effects of the tsunami on the coast of Brazil. The model town is Rio de Janiero. Sea begins to reverse some people will notice that. Residents begin to warn others of the impending tidal wave of themselves do tend to run inland. Some ignore warnings remain indifferent. Wave approaching, people fleeing in panic. Elderly citizens remains as the home because they do not know about anything. Wave hits full force, swept the first dwelling closest to the coast several residents who be enough shelter. The aim is to determine how strong the tsunami must be to destroy as many buildings many people will be at risk of life. The estimated death toll will be around 60-70%. --Hanka (talk) 23:45, 16 December 2016 (CET)

Wow, this is very ambitious. I would really love to see the result, but I am afraid, it will be incredibly complex to do it this way. Too many factors in the game: the landscape, buildings, people, their panic, how do they behave, physics of the water, etc. etc. etc. Try to narrow the problem down and describe how exactly should the solution look like. Then it could be really good. Tomáš (talk) 14:03, 18 December 2016 (CET)

I will work with the following parameters: Wave - speed of 500-700 km/h, the height of 15-30 meters, width of 30-50km. The density of buildings of about 3milion. Population 6,5milion, the area of 1260km2. The result will be 80% of the cities covered by water, the loss of life of 70%, but this may be varied depending on the diversity of the parameters of wave. --Hanka (talk) 17:31, 18 December 2016 (CET)

Ok, how it will actualy look like? How you will deal with spatial characteristics? Perhaps a small sketch would help. The idea is really appealing, I just want to be sure that you have an exact idea, how you want to sort it out as it could be a potential headache. Tomáš (talk) 00:34, 22 December 2016 (CET)

Courier is coming to town (Maria Alekseeva, Mashal)

There's a small village with certain number of houses and it doesn't have its own local delivery service, thus once a week a delivery person drives up to this village from a bigger city. Each time this person has a different delivery list and his route is based around that list. If for some reason nobody was able to receive a package at the time that the delivery person was there, he has to come back to the same address until the package is delivered. The simulation in NetLogo will be creating the perfect route for the delivery person based on A* search algorithm, key parameters are the number of times, when delivery person must return to the house and the number of houses.

But even though I'm dying to make search algoritmus in NetLogo, I'm not quite sure, that this simulation is fine or if it's agent-based... That's why I have one more and I'm asking you to recommend me, which one is better

Depression spreading on social media There is certain number of people and everyone has several parametrs: aktivity (how often writes smth on social media, const), mood (in per sent, changing parameter, if p>50% man is "optimist", if p<50% man is "pessimist"). Optimist can write only sad posts, pessimist writes only happy posts Man can read different posts from people in certain radius and if the post is happy he gets happier and if the post is sad he gets sad. And than he writes his own posts. This simulation will show how one pessimist with a big activity can make the whole society sad. I'm not sure if it can be developed in NetLogo< because i need counter for the popularity parameter, but i can do it in C++

The former is definitely better. Regarding the fit of NetLogo for the problem, it is doable and among all four platforms we use, it is the best option for the task indeed. Tomáš (talk) 14:18, 18 December 2016 (CET)
Fine, so I'll do my first version in NetLogo--Mashal (talk) 17:26, 18 December 2016 (CET)
Approved. Tomáš (talk) 00:35, 22 December 2016 (CET)

Ocean currents influencing the temperature of the Earth (Stanislav Vedra, xveds02)

Environment: NetLogo

Variables are a certain amount of units of warm water (e.g. caused by the sun) and cool water (e.g. caused by an iceberg) with the same size spread in the environment. Warm and cool water are moving until they reach each other which set off the current. The temperature of the current is decided by 50% chance. The size of the current is a total of the variables. As well as waters merge, currents do too. With the same size of currents, the rules are the same (50% chance of the temperature). With different size, the winner is the bigger one and takes everything. In both cases, the size of the new current is a total of previous ones.

Variables and currents are mixing until one current will be left. And its temperature means if the Earth is warming or cooling down, which is the goal of the simulation.

Ok, the problem sounds interesting. How you will deal with real data? Tomáš (talk) 14:22, 18 December 2016 (CET)
The problem with influencing the Earth temperature by ocean currents is very complex not mention it is not all discovered yet, there is not enough data from the past. Therefore, I took this little part of it and the provided data will be still an estimation. The simulation is mainly for illustrative purposes. Although I will try to get to the reality as close as I can, alternatively add scroll bar/input box to choose desired data, but I would rather not promise anything realistic. I hope I answered you question? --Xveds (talk) 18:14, 18 December 2016 (CET)
I am sorry, but in such a case, we cannot do it. You will have to come to a conclusion. The task is not about the simulation. Simulation is just an instrument leading to a report. If you cannot "promise anything realistic", we have no chance how to evaluate whether your conclusions make sense or not. You can come with virtually anything and there will be no way how to verify it. In science, it is called "falsifiability". I am afraid that your suggestion is not falsifiable, hence it cannot be accepted. Please, try to either reformulate it somehow, narrow the problem down, or simply find something completely different. Tomáš (talk) 00:41, 22 December 2016 (CET)

Traffic simulation(Jakub Vitvar, xvitj27)

Software: NetLogo

The enviroment would be graph with variable amount of nodes(cities). Agents would simulate traffic in this enviroment. Turtles would travers the graph from their set home node to their target node and back. They would find their way by using A* algorithm. Every edge would have set distance and max speed.

Simulation would create road closures. Agents would try to avoid these closures and traffic jams based on their set variables (looking for shortest or fastest way). Simulation is going to show weak points of this traffic system during traffic problems. Xvitj27 (talk) 16:40, 19 December 2016 (CET)

Good, traffic simulations are among the most popular and quite useful for ABM, however, in your case, it looks like a bit general. You should either reformulate it in order to solve a particular traffic problem (like e.g. the influence of "no passing rule" on traffic density or something like that), or, perhaps easier, simulate precisely a particular spot (like e.g. a particular junction, a particular tunnel, etc.). Tomáš (talk) 00:46, 22 December 2016 (CET)

So if I pick for example first 30 kilometers of D1 and simulate the exits and neighbouring nodes(cities). The simulation would show effects of accidents and road restrictions on D1 on the close cities. Would that be okay? Xvitj27 (talk) 14:17, 22 December 2016 (CET)

Approved. Don't forget to obtain real data. Fortunately, it should be easy for this particular example.