Flexible model structures for discrete choice analysis.

Econometric discrete choice analysis is an essential component of studying individual choice behaviour and is used in many diverse fields to model consumer demand for commodities and services. This chapter provides an overview of the motivation for, and structure of, advanced discrete choice models derived from random-utility maximization. The discussion is intended to familiarize readers with structural alternatives to multinomial logit (MNL) and to the models discussed in Chapter 13 (see ITRD E109039). Before proceeding to a review of advanced discrete choice models, the assumptions of the MNL formulation are summarized. This is useful since all other random-utility maximizing discrete choice models focus on relaxing one or more of these assumptions. The first assumption is that the random components of the utilities of the different alternatives are independent and identically distributed (IID) with a type I extreme-value (or Gumbel) distribution. The second assumption of the MNL model is that it maintains homogeneity in responsiveness to attributes of alternatives across individuals (i.e., an assumption of response homogeneity. The third assumption is that the error variance-covariance structure of the alternatives is identical across individuals (i.e., an assumption of error variance-covariance homogeneity. In this chapter the focus is on two classes of discrete choice models that relax one or more of these assumptions. The first class of models (here called "heteroscedastic models") are relatively restrictive; they relax the identically distributed (across alternatives) error term assumption, but do not relax the independence assumption (part of the first assumption above) or the assumption of response homogeneity (second assumption above). The second class of models (here called "flexible structure models") are very general; models in this class are flexible enough to relax the independence and identically distributed (across alternatives) error structure of the MNL as well as to relax the assumption of response homogeneity. The third assumption is not considered in detail in this chapter, since it can be relaxed within the context of any given discrete choice model by parameterizing appropriate error structure variances and covariances as a function of individual attributes.