STATS 783 Simulation and Monte Carlo Methods

Below description edited in year: 2010

Points: 15

Credit: Examination 50%, Assignments and projects 50%

For Advice: Yong Wang (Email: yong.wang@auckland.ac.nz | extn: 84700)

Taught: First Semester City

The aim of STATS 783 is to give a practical introduction to modern simulation and Monte Carlo techniques. These techniques use computer-generated random numbers to simulate real situations and to solve problems where mathematical analysis is hard or intractable. The course is intended to be practical, with a focus on methodology and use of computers for problem-solving. Theory needed is restricted to a minimum level.

Topics studied include: introduction to statistical inference and computing, random number generation (uniform; non-uniform; vector), Monte Carlo integration, variance reduction techniques (correlated sampling; antithetic variates; control variables; conditional expectation; stratified sampling; importance sampling), data resampling techniques and applications (bootstrap; jackknife; cross-validation; prediction error estimation; model selection; nonparametric regression; bootstrap confidence intervals; resampling hypothesis tests), introduction to Markov chains (discrete/continuous state), Monte Carlo optimization (simulated annealing; recursive integration; genetic algorithms; Monte Carlo EM), algorithms and applications of Markov chain Monte Carlo (Metropolis-Hastings algorithm; Gibbs sampler; completion construction; data augmentation; random effects models).

Disclaimer:
Although every reasonable effort is made to ensure accuracy, this information for the course year (2011), is provided as a general guide only for students and is subject to alteration. All students enrolling at the University of Auckland must consult its official document, the University of Auckland Calendar, to ensure that they are aware of and comply with all regulations, requirements and policies.