Discrete/continuous choice models and consumer heterogeneity

This dissertation consists of three chapters on demand estimation using discrete and discrete/continuous choice models. In all papers I address the identification of heterogeneity in preferences using random parameter specifications. From a policy perspective, preference heterogeneity is important in the design of policy interventions targeting groups with different behavior patterns and, in a political economy context, enables us to identify winners and losers from these interventions, and to deal with them appropriately.;Framed in a Lancasterian demand model, where demand depends on the attributes of the goods, the first chapter entitled "Taste Indicators and Heterogeneous Revealed Preferences for Congestion in Recreation Demand" presents a method combining taste indicators and revealed preferences to estimate a distribution of preferences for attributes of a commodity. This work is motivated by the fact that results of a traditional Mixed Logit (ML) model are sometimes inconsistent with people's attitudes towards those attributes. I include these attitudes into the estimation process and I utilize them to describe the distribution of preferences in the sample. I apply this procedure to estimate a distribution of taste for congestion in a recreation demand model. While the traditional ML model suggests that around 60% of the sample likes crowding, my estimations show that nobody in the sample likes this attribute of the sites. A correct estimation of the distribution of tastes in the population may contribute to correctly assess the benefit or cost of some policy intervention.;The second chapter, "Functional Forms in Discrete/Continuous Choice Models with General Corner Solutions", presents a new utility model serving as the basis for modeling discrete/continuous consumer choices with a general corner solution. The new model involves a more flexible representation of preferences than what has been previously used in the literature and, unlike most of it, the model is not additively separable. This functional form can handle richer substitution patterns such as complementarity and substitution among goods. I focus in part on the Quadratic Box-Cox utility function and examine its properties from both theoretical and empirical perspectives. I also identify the significance of the various parameters of the utility function, and demonstrate an estimation strategy that can be applied to demand systems involving both a small and large number of commodities.;Finally, the chapter three entitled "Quality Effects on the Demand for Broadway Shows" presents an estimation of live theatre demand using a discrete choice model commonly used in industrial organization applications with aggregate data. The focus of this paper is the identification of the effect of quality on attendance to a Broadway show. Quality is defined as the number of Tony awards won by each show. A positive correlation between admission price and show's quality has produced upward sloping demand curves in previous studies in the economics of the Arts, contradicting economic theory. The application in this chapter uses an unique panel dataset created by matching weekly Broadway show attendance, theatre gross revenue, and capacity figures with show attributes for all plays performing between December 1995 and June 2003. Using a discrete choice model with instrumental variables the results show Broadway show demand slopes down, the Tony award has a positive impact on demand, individuals like Musicals more than Plays, and Hollywood actors draw bigger crowds. Obtaining consistent estimates of the demand for Broadway shows opens several avenues for future work in this topic such as market concentration and competition among firms and survival analysis for plays.