Demand Models for Market Level Data

In this paper I propose new continuous and discrete choice demand models. To do so, I note that existing demand models from the continuousand discrete-choice demand literatures each have distinct advantages for taking to market-level data. I attempt to build models with the union of the good properties of each literature. For instance, the continuous choice models are more appropriate for disaggregated data than popular demand models such as the Translog or Almost Ideal Demand Systems since they can be estimated even when products enter or exit the market during the sample period. Variation in the observed set of products can then be used to help identify substitution patterns, in a way recently made popular in the discrete choice demand literature. I propose a continuous choice model that can use this important source of identifying data variation. I then propose a discrete-choice demand model that is provides a flexible functional form in the sense Diewert (1974). It is a particular member of the GEV class of models developed by McFadden (1981). In each case I argue that developing highly parameterized demand models and then mapping those parameters down to be functions of product characteristics (following Pinkse, Slade, and Brett (2002)) provides a simple, coherent, and data driven method for estimating very rich demand systems. Approaching aggregate demand models in this way is computationally much less demanding than using a random coefficient approach since it does not require the use of simulation estimators. Along the way, I propose a generalization to the contraction mapping developed in Berry, Levinsohn, and Pakes (1995) for discrete choice models based on the observation that the Expected maximum utility function for the GEV class of models is a super-modular function. J.E.L. Classification: D0, D11, D12, C51