演講者簡介 : Che-Lin Su got his Ph.D. in Operations Research from Stanford University in 2005. He is currently serving as Assistant Professor of Operations Management at Booth of College at The University of Chicago. His research field is in structural estimation with applications in operations, economics, and quantitative marketing, incentive models with applications to executive compensation design and nonlinear pricing, and optimization and computational economics.
演講摘要 : A pure characteristics model (PCM), studied in Berry and Pakes (2007), is a class of discrete-choice random-coefficients demand models in which there is no idiosyncratic logit error term in a consumer's utility. The absence of the logit error term leads to a nonsmooth formulation of the predicted market share equations. As a result, inverting the market share equations for the unobserved product characteristics and estimating the model becomes computationally infeasible using nested-type algorithms. In this talk, we formulate the GMM estimation of a pure characteristics model as a mathematical programs with complementarity constraints (MPCC), where the equilibrium constraints characterize consumers' decisions. We also present the formulation of the PCM model with supply-side equations and discuss the computational challenges in estimating these models with the MPCC approaches.
