Stochastic Simulation

Autumn semester 2018

General information

Lecturer Dr. Fabio Sigrist
Assistants Niklas Pfister
Lectures Tue 14-17 ML F 36 >>
Exercises Tue 16-17 (≈ biweekly) ML F 36 >>
Course catalogue data >>

Course content

Examples of simulations in different fields (computer science, statistics, statistical mechanics, operations research, financial mathematics). Generation of uniform random variables. Generation of random variables with arbitrary distributions (quantile transform, accept-reject, importance sampling), simulation of Gaussian processes and diffusions. The precision of simulations, methods for variance reduction. Introduction to Markov chains and Markov chain Monte Carlo (Metropolis-Hastings, Gibbs sampler, Hamiltonian Monte Carlo, reversible jump MCMC).

Announcements

  • September 14th, 2018:
    Beginning of lecture: Tuesday, 18/09/2018. (Exercises start on 25/09/2018 at 16:15).

Course materials

  • Lecture Notes
  • Slides used in the course will be available in due time. Exercises as well as solutions will also be provided.
  • More information available in Organization sheet


Week Date Lecture/Exercise Topic
1 18/09 L-L-L Introduction, distribution of estimators, simulation in Bayesian statistics
2 25/09 L-L-E Simulation in statistical mechanics and operations research
3 02/10 L-L-L Simulation in financial mathematics, MC integration, accuracy of MC methods
4 09/10 L-L-E Generating uniform random variables
5 16/10 L-L-L Quantile transformation, rejection sampling, relations between distributions, permutations
6 23/10 L-L-E Importance sampling , simulation of stochastic differential equations
7 30/10 L-L-E Variance reduction: antithetic variables, control variates, importance sampling
8 06/11 L-L-L Quasi Monte-Carlo, introduction MCMC, basics of Markov chains
9 13/11 L-L-E Metropolis-Hastings algorithm
10 20/11 L-L-L Independence sampler, random walk Metropolis, componentwise modification, Gibbs sampler
11 27/11 L-L-E Hamiltonian Monte Carlo, introduction to reversible jump MCMC
12 04/12 L-L-L Reversible jump MCMC
13 11/12 L-L-E Accuracy of MCMC approximations
14 18/12 L-L-L reserve / buffer

Exercise classes

Exercises will be held roughly bi-weekly, but on an irregular schedule. The statistical software package R is recommended for solving the exercises. The exercises will take place at the specified date from 16.15-17.00 in the same place as the lectures.

Series and solutions

Only phd students who are not taking the exam need to hand-in at least 5 well-solved exercises in order to get credits for the course. All other students do not need to hand in the exercises. You can hand in your solution during the class or by email until the designated date and will receive some feedback in due time.

Date Exercises Solutions Due date
25/09 Series 1 Solution 02/10
09/10 Series 2 Solution 16/10
23/10 Series 3 Solution 30/10
30/10 Series 4 Solution 06/11
13/11 Series 5 Solution 20/11
27/11 Series 6 Solution 06/12
11/12 Series 7 Solution 18/12

Introductory books

  • G. S. Fishman, A First Course in Monte Carlo. Thomson Brooks/Cole, 2006.
  • S. M. Ross. Simulation. Academic Press, 2012 (5th edition).

Books on a similar level as the course

  • Ch. Robert and G. Casella. Introducing Monte Carlo Methods with R. Springer Science & Business Media, 2009.
  • Ch. Robert, G. Casella. Monte Carlo Statistical Methods. Springer 2004 (2nd edition).
  • S. Asmussen, P. W. Glynn, Stochastic Simulation, Algorithms and Analysis. Springer, 2007.
  • P. Glasserman, Monte Carlo Methods in Financial Engineering. Springer 2004.
  • B. D. Ripley. Stochastic Simulation. Wiley, 1987.
  • W. R. Gilks, S. Richardson, D. J. Spiegelhalter. Markov Chain Monte Carlo in Practice. Chapman & Hall, 1996.
  • S. Brooks, A. Gelman, G. Jones, and X.L. Meng, eds. Handbook of Markov Chain Monte Carlo. CRC press, 2011.