Analysis using generalized linear models and its applied computation with R

Abstract

Generalized linear models (GLMs) are introduced by Nelder and Wedderburn (see [8]). As an extension of normal linear regression for a single dependent variable, GLMs are widely used to do regression modeling for non-normal data with a minimum of extra complication. Unifying various other statistical models under one framework, GLMs develop a general algorithm for maximum likelihood estimation in all these models. GLMs are flexible enough to include a wide range of common situations, but at the same time allow most of the familiar ideas of normal linear regression to carry over. Furthermore, the link function which provides the relationship between the linear predictor and the mean of the distribution function does not have to be linear. Taking advantage of statistical software R (see [3]), efficiency of GLMs analysis in different datasets is greatly improved and diagnostics become visible.