Skip to main content

Trinity College Dublin, The University of Dublin

Menu Search

Module Descriptor School of Computer Science and Statistics

Module CodeST3456
Module Short TitleSimulation methods
Semester TaughtHilary
Contact HoursLecture hours: 33 
Module PersonnelDr. Bernardo Nipoti
Learning Outcomes

Students will have ability: to devise suitable simulation methods for generating random numbers from a given probability distribution to use the sampled random numbers in order to estimate quantities of interest or evaluate integrals to assess the quality of the generated sample via diagnostic tools

Learning Aims

The course will introduce simulation methods with particular focus on Monte Carlo methods based on Markov Chains. Theoretical justification of basic concepts will be provided together with examples where such methods are applied to problems arising with statistical practice.

Module Content

Specific topics addressed in this module include:

  • Random variable generation: transformation methods, accept-reject methods
  • Monte Carlo integration and importance sampling
  • Essentials of Markov Chains Markov Chain Monte Carlo Methods: Metropolis-Hastings and Gibbs sampler
Recommended Reading List

Main text:

Rubinstein and Kroese, “Simulation and the Monte Carlo method”, 2nd edition, Wiley 2008.

Useful references:

Devroye, “Non-uniform random variate generation”, Springer 1986. (the book is freely available on the author’s webpage)

Robert & Casella, “Monte Carlo statistical methods”, Springer 2004.

Additional material might be provided during the course.

Module PrerequisitesFamiliarity with basic concepts in probability and statistics.
Assessment Details

Exam (100%), 2 hours of time

Module Website3456
Academic Year of Data2016/17