Categorical data analysis

an overview of explanatory disrete data methods and disaggregate choice models

Abstract

This paper focuses attention on recent advances in the area of discrete data analysis and
probabilistic choice models. The emphasis is put on the use of these methods in an explanatory context. The first part of the paper provides a survey of recently developed explanatory discrete (categorical) data methods (linear logit models, log-linear models, e.g.). The second part is concerned with explanatory discrete choice models (multinomial logit models, generalized extreme value models, e.g.). The paper concludes with a discussion of promising new research directions in the area of discrete data and choice analysis.