vix.ing · top · new · best · stats · spec

Semiparametric Estimation of a CES Demand System with Observed and\n Unobserved Product Characteristics

2015/11/17 by Alı Hortaçsu, Hortacsu, Ali, Joonhwi Joo +1
Business, Management and Accounting · Economics, Econometrics and Finance · #Applications (stat.AP) #Consumer Market Behavior and Pricing #Economics of Agriculture and Food Markets #FOS: Computer and information sciences #Merger and Competition Analysis

paper · pdf · doi:10.48550/arxiv.1511.05600

openalex publication_date 2015/11/17 · openalex created_date 2019/07/30 · openalex updated_date 2026/07/28

Abstract

We develop a characteristics based demand estimation framework for the\nMarshallian demand system obtained by solving a budget-constrained constant\nelasticity of substitution (CES) utility maximization problem. From our\nMarshallian CES demand system, we derive the same market share equation of\nBerry (1994); Berry, Levinsohn, and Pakes (1995)'s characteristics based logit\ndemand system. Our CES demand estimation framework can accommodate zero\npredicted and observed market shares by conceptually separating the\nwhether-to-buy decision and how-much-to-buy decision. Furthermore, the\nestimator we suggest allows a tractable semiparametric estimation strategy that\nis flexible regarding the distribution of unobservable product characteristics.\nWe apply our framework to scanner data on cola sales, where we show estimated\ndemand curves can be upward sloping if zero market shares are not accommodated\nproperly.\n

Citations

Related