On generalizing the multiple discrete-continuous extreme value (MDCEV) model

The overall goal of the dissertation is to contribute to the growing literature on multiple discrete-continuous (MDC) choice models. In MDC choice situations, consumers often encounter two inter-related decisions at a choice instance – which alternative(s) to choose for consumption from a set of available alternatives, and the amount to consume of the chosen alternatives. In the recent literature, there is increasing attention on modeling MDC situations based on a rigorous underlying micro-economic utility maximization framework. Among these models, the multiple-discrete continuous extreme value MDCEV model (Bhat, 2005, 2008) provides a number of advantages over other models. The primary objective of this dissertation is to extend the MDCEV framework to accommodate more realistic decision-making processes from a behavioral standpoint. The dissertation has two secondary objectives. The first is to advance the current operationalization and the econometric modeling of MDC choice situations. The second is to contribute to the transportation literature by estimating MDC models that provide new insights on individuals’ travel decision processes.

The proposed extensions of the MDCEV model include: (1) To formulate and estimate a latent choice set generation model within the MDCEV framework, (2) To develop a random utility-based model formulation that extends the MDCEV model to include multiple linear constraints, and (3) To extend the MDCEV model to relax the assumption of an additively separable utility function. The methodologies developed in this dissertation allow the specification and estimation of complex MDC choice models, and may be viewed as a major advance with the potential to lead to significant breakthroughs in the way MDC choices are structured and implemented. These methodologies provide a more realistic representation of the choice process. The proposed extensions are applied to different empirical contexts within the transportation field, including participation in and travel mileage allocated to non-work activities during various time periods of the day for workers, participation in recreational activities and time allocation for workers, and household expenditures in disaggregate transportation categories. The results from these exercises clearly underline the importance of relaxing some of the assumptions made, not only in the MDCEV model, but in MDC models in general.