CMStatistics 2020: Start Registration
View Submission - CFE
Title: Consumer theory with non-parametric taste uncertainty and individual heterogeneity Authors:  Christopher Dobronyi - University of Toronto (Canada) [presenting]
Christian Gourieroux - University of Toronto and CREST (Canada)
Abstract: Two new classes of non-parametric random utility models for demand systems are introduced. In each class, individual-level heterogeneity is characterized by a distribution $G$ over taste parameters, and heterogeneity across consumers is introduced by means of a distribution $F$ over the distributions $G$. Demand is non-separable and heterogeneity is infinite-dimensional. Each class allows for corner solutions. We present two distinct frameworks for model estimation: (i) a Bayesian framework in which $F$ is known, and (ii) a hyperparametric framework in which $F$ is a member of a parametric family. We use a panel of scanner data to illustrate our methods in an application to the consumption of alcohol.