Computational statistics is the interface between statistics and computing. We will cover algorithms for solving statistical problems for model fitting, prediction/generalization, and uncertainty quantification. We will talk about how to use optimization and other modern computational techniques to develop algorithms for various types of data. Topics include linear and non-linear regression, splines, density estimation and kernel methods, Gaussian mixture models and EM algorithm, hidden Markov models, Monte Carlo Markov chain (MCMC), and bootstrap.

This course provides academic credit for independent thesis research conducted under the supervision of a Georgia Tech faculty advisor.

Introduction to probability, probability distributions, point estimation, confidence intervals, hypothesis testing, linear regression, and analysis of variance. Crosslisted with MATH 3770 and CEE 3770. Also, credit not awarded for both ISYE 3770 and MATH 3670.