Simulation 30E00400

GD Star Rating
loading...

Was the course worth attending? Were the lectures useful? What questions did the exam have? Give us your honest opinion on the course!

3 Replies to “Simulation 30E00400”

  1. I entered the course with vague memory of statistics basics. When the professor was listing the recommended prerequisites, I met very few of those. I knew I would have a hard time, and I did. However, if you are prepared to put in the work you shouldn’t have too many issues. If your diffentential and integral calculus, as well as memory of basic statistics, are on point, this course should not be too difficult.

    The content is useful and interesting. There is no longer a course project, instead there are more exercise sets, 8 in total. The exam is no joke, you need to know your stuff if you want to get a good grade. Even if you know how to answer the questions without thinking about them too much, you will not leave the exam early. I had no time to check all my answers even though I knew immediately how to solve those.

    Bonus tip is to be prepared to answer the question 4) listed by Anti-Tantta. We had a similar question and it wasn’t even the last one.

  2. In 2016, the course was pretty much the same as described by Anti-Tantta. The content is quite useful, but its a rough ride. You get 6 credits (if you pass) but you work for 12, or more. The exercise sets expect you to know or learn a lot of stuff not covered in class or the slides, and the last two exercise sets are quite brutal. Do not take this course if you have other commitments in the first period such as a job or another demanding course. You won’t have a good time.

  3. The topics covered in the course are very useful, and it teaches a whole new approach to problem solving if you aren’t familiar to simulating stuff. I never thought about using Excel in the way we did in this course. Excel is used to build simulation models since everyone is familiar with Excel, and with it students are pretty well able to understand the underlying logic in making simulation models. I recommend the course to people who want to learns something useful in school they can use in real life.

    That being said, the course is very burdensome. It lasts for two periods. In the first period there are 4-5 weekly exercises, each requiring about 3-4 days of work to get them done decently. I recommend you do them since the exercises account for 20% of the points you get for the course. You will need them.

    The second period is dedicated for doing a project. You can do it alone or with a partner. You should try to do it very well, it accounts for 30% of the points for the course. You will need them.

    There are several ways to agglomerate bonus points. Attending the practice session brings you 1 point per session, some of the homework give bonus points, you get bonus points for evaluating other teams’ presentations, and for reading through other groups project reports and evaluating them. There is more bonus work available than you will bother doing, but try to gather as many bonus points as you can. You will need them.

    Then there is the exam. If you haven’t figured out by now, it is quite hard and accounts for 50% of the grade. It had 4 questions (each worth 20 points), something like this:

    1) Answer the following questions (4 points each)
    – Describe what are the two most important distributions in simulation and why they are important.
    – What issues affect the accuracy of a simulation
    – Explain what the central limit theory is. Include as much mathematical explanation as possible.
    – Describe what is a random walk with drift
    – Explain the acceptance/rejection method

    2) Explain how to simulate values for the following distributions/cases. Then generate a value for the random variable X in each case assuming that a random number U = 0.47 was generated for the simulation. (5 points each)
    – X~LogN(10, 2^2)
    – X is the earliest arrival for an object following a Poisson distribution [or something like that, I don’t remember exactly how this question was formed]
    – Leo is practicing soccer kicks. His probability of scoring is 0.7. X is the result after Leo has made three consecutive kicks. What is the simulated value for X?
    – Same as previous except now X is the amount of kicks Leo makes for scoring his first goal.

    3) a) f(x) = some second order polynomial. Show that it is a pdf. How would you simulate random numbers from this distribution? (10 points)

    3) b) f(x) = h(2-x)(3-x) when 0≤x≤2, f(x)= 0 otherwise. Determine the value of h so that f(x) is a pdf. Show how you would simulate values from this distribution. (10 points)

    4) This one was an insane, ridiculous killer. The problem description alone is very long and hard to understand. I’ll just sort of try to describe what it said since I can’t recall properly. Basically you have to simulate two stock prices whose returns both follow a logarithmic return. The question says you should utilize Ito’s lemma. The logarithmic returns are also correlated with each other with a certain coefficient. You are given yearly variances for X and Y (the question asks them on a monthly basis). You are also told ε1 and ε2 have a covariance of 0.2. You are asked also to simulate X(t+1) and Y(t+1). You are given some values for X(t) and Y(t), and two random numbers (0.66 and 0.47). At the bottom of the exam paper it says ”remember to :)”. I sure as hell wasn’t smiling. (20 points)

Vastaa

Sähköpostiosoitettasi ei julkaista. Pakolliset kentät on merkitty *

This site uses Akismet to reduce spam. Learn how your comment data is processed.