Skip to main content

Generalized Linear Models: Course Outline

In statistics, the generalized linear model (GLM) is a flexible generalization of ordinary linear regression that allows for response variables that have error distribution models other than a normal distribution.

E-Books (Full Text)

Course Outline

  • Logit model for dichotomous data with  single  and  multiple explanatory variables ,
  • maximum likelihood (ML)  estimation,
  • large sample tests about parameters ,
  • good of fit , 
  • analysis  of deviance ,
  • variable selection ,
  • extension to polytomous data ,
  • introduction to Poisson regression.
  • Linear models for two and three dimensional contingency tables:
  • interpretation of parameters,
  • comparison with ANOVA and regression.
  • ML estimation of parameters.
  • Likelihood ratio tests for various hypotheses,
  • including independence,
  • marginal and conditional independence,
  • partial association .
  • Models with quantitative levels.
  • Nonparametric regression and generalized linear models.
  • Interpolating and smoothing splines for simple regression.
  • Use of cross-validation Applications to Logistic and Poisson regression.
  • Additive models and generalized additive models.   

E-Books (Full Text)