STAT 109 Generalized Linear Models

Reviews and extends ideas of multiple regression analysis to a wider class of models—Generalized Linear Models (GLMs)—involving the relationship between a response and one or more explanatory variables. Models for the analysis of quantitative and qualitative responses, including multiple regression, analysis of variance, covariance models, binomial models for binary responses (including logistic regression and probit models), and Poisson models for count data. Case studies drawn from social, engineering, and life sciences. (Formerly AMS 174.)