Probability Workshop 2019

Title: Monte Carlo Methods
Facilitator: Aaron Smith, University of Ottawa;

Dr. Smith is an Assistant Professor in the Department of Mathematics and Statistics at the University of Ottawa. He works primarily in the area of applied probability, with a focus on Markov chain Monte Carlo and related methods from computational statistics or statistical physics. 

Workshop Description: Monte Carlo (MC) methods, the practice of approximately simulating from a probability distribution in order to understand it, are popular throughout statistics and the sciences. Although there are good software packages for many popular MC algorithms, there are no "black box" methods that work well for all realistic problems. As a result, many statisticians find that they must implement their own algorithm variants in order to get good results. This workshop will provide an introduction to various MC methods. The workshop is aimed at individuals who have limited experience with MC methods. The workshop will provide: (a) a general introduction to popular Markov Chain MC methods and some associated software implementations, and (b) longer case studies of somewhat-realistic statistical problems for which "default" methods perform poorly. Although good implementations are very important when using MC methods, it is not practical to debug complicated code in real-time during a workshop. For this reason, the workshop will emphasize algorithm development and concepts.