STAT 454 - Bayesian Statistics

Bayesian statistics, an alternative to the traditional frequentist approach taken in our other statistics courses, is playing an increasingly integral role in modern statistics. The Bayesian philosophy is natural, allowing us to formally balance data with our prior knowledge, and updating this knowledge as more data come in. It answers natural questions. It can shine in settings where frequentist “likelihood” methods break down. And it is becoming increasingly popular with the availability of computing tools necessary to its implementation. This course explores the Bayesian approach to statistical analysis, Bayesian computing, and both sides of the frequentist versus Bayesian debate. Topics include Bayes’ Theorem, prior and posterior probability distributions, Bayesian regression, Bayesian hierarchical models, and an introduction to Markov chain Monte Carlo computing techniques. Prerequisite(s): STAT 155 and MATH 354. Offered occasionally. (4 Credits)