Description

This course is a practical introduction to modern computational methods that can be used to solve complex statistical and machine learning problems. The topics covered include basic optimization techniques, permutation tests, bootstrapping, cross-validation, the Expectation-Maximization (EM) algorithm, and Monte Carlo algorithms,  including importance sampling, Markov Chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC). The course will consider use of these methods in applications as well as discuss some of the methods’ theoretical properties. Basic proficiency in R, Python or another language suitable for statistical computing is assumed.

 

 References

   1) Computational Statistics by G.H. Givens and J.A. Hoeting. Wiley, 2005

   2) Monte Carlo Strategies in Scientific Computing by Jun S.Liu. Springer. 2004.

   3) Handbook of Markov Chain Monte Carlo, editors S. Brooks, G.  Jones and Xiao-Li Meng.  Chapman and Hall. 2011.