Models

• We will look at a number of related ways of modelling categorical data
• Loglinear models: model the structure of multi-dimensional tables of categorical variables - ``association'' models no dependent variable
  • 2-dimensional tables
  • Higher dimensions
  • Special topics such as square tables or ``matched pairs''
  • Models for ordinal variables
• Grouped logistic regression: modelling tabular data with a binary dependent variable
• Individual-level logistic regression: micro-data with a binary dependent variable, continuous covariates possible
• Multinomial logistic regression: dependent variable with more than two categories
• Ordinal logistic regression: dependent variable with more than two categories, which are ordered