Simulation error and numerical instability in estimating random coefficient logit demand models

A-Tier
Journal: Journal of Econometrics
Year: 2025
Volume: 247
Issue: C

Authors (4)

Brunner, Daniel (not in RePEc) Heiss, Florian (not in RePEc) Romahn, André (Heinriche-Heine-Universität Dü...) Weiser, Constantin (not in RePEc)

Score contribution per author:

1.005 = (α=2.01 / 4 authors) × 2.0x A-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

The nonlinear GMM-IV estimator of Berry, Levinsohn and Pakes (1995) can suffer from numerical instability resulting in a wide range of parameter estimates and economic implications. This has been reported to depend on technical details such as the choice of the optimization algorithm, starting values, and convergence criteria. We show that numerical approximation errors in the estimator’s moment function are the main driver of this instability. With accurate approximation, the estimation approach is well-behaved. We provide a simple method to determine the required number of simulation draws.

Technical Details

RePEc Handle
repec:eee:econom:v:247:y:2025:i:c:s0304407625000077
Journal Field
Econometrics
Author Count
4
Added to Database
2026-01-29