Computational Statistics

Catalog description

3 Lecture Hour(s) 0 Lab Hour(s) This course introduces students to various computationally intensive statistical techniques. Topics will include numerical optimization for statistical inference (gradient-based optimization, the Expectation-Minimization (EM) algorithm, and Fisher scoring), random number generation, resampling methods such as the bootstrap, permutation and randomization tests, cross-validation, Markov Chain Monte Carlo techniques (Gibbs sampling and Metropolis-Hastings algorithm), and nonparametric curve fitting. Students will learn to apply these techniques to solve data science problems using the statistical software R.