Estimation of random coefficients logit demand models with interactive fixed effects

2026-05-01

AI Summary

This study addresses the limitations of traditional random coefficients logit models, which struggle with endogeneity arising from unobserved product characteristics correlated with prices and fail to account for cross-product and cross-market share persistence. Within the Berry–Levinsohn–Pakes (BLP) framework, the paper innovatively introduces a factor-structured interactive fixed effects specification that permits arbitrary correlation between unobserved attributes and prices, thereby mitigating endogeneity while capturing dynamic market share patterns. To implement this approach, the authors develop a computationally efficient two-step least squares–minimum distance (LS–MD) estimator and demonstrate its strong finite-sample performance through Monte Carlo simulations. An empirical application to U.S. automobile market data confirms both the theoretical rigor and practical feasibility of the proposed method.

Abstract

We extend the Berry, Levinsohn and Pakes (BLP, 1995) random coefficients discrete-choice demand model, which underlies much recent empirical work in IO. We add interactive fixed effects in the form of a factor structure on the unobserved product characteristics. The interactive fixed effects can be arbitrarily correlated with the observed product characteristics (including price), which accommodates endogeneity and, at the same time, captures strong persistence in market shares across products and markets. We propose a two-step least squares-minimum distance (LS-MD) procedure to calculate the estimator. Our estimator is easy to compute, and Monte Carlo simulations show that it performs well. We consider an empirical illustration to US automobile demand.

Problem

• random coefficients
• demand models
• interactive fixed effects
• endogeneity
• unobserved product characteristics

Innovation

• interactive fixed effects
• random coefficients logit
• factor structure
• endogeneity
• least squares-minimum distance