Information matrix tests for multinomial logit models

C-Tier
Journal: Economics Letters
Year: 2025
Volume: 247
Issue: C

Score contribution per author:

0.335 = (α=2.01 / 3 authors) × 0.5x C-tier

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

Abstract

We obtain simple and intuitive expressions for the information matrix test for the multinomial logit model that resemble a multivariate heteroskedasticity test à la White (1980) applied to the conditionally demeaned value of the outer product of the generalised residuals.

Technical Details

RePEc Handle
repec:eee:ecolet:v:247:y:2025:i:c:s0165176525000175
Journal Field
General
Author Count
3
Added to Database
2026-01-25