Keren Fang

University of California, Riverside · Economics · 2026–27 job market

Fields

Econometrics, Theory, Industrial Organization

Job market paper

Inference for Moment Inequalities with Nuisance Functions

This paper proposes a conceptually novel semiparametric model formulated by moment inequalities, which involve partially identified finite-dimensional parameters of interest and point identified infinite-dimensional nuisance parameters. We propose a reparametrized GMM condition to obtain the variance estimator by a corresponding influence function. We partially identify the parameters of interest through their confidence set. We establish the uniform bootstrapping consistency and the uniform asymptotic size control for the constructed confidence set. We illustrate the method by a structural entry game model with partially identified state-dependent interaction effects, where the true conditional choice probabilities serve as the nuisance functions.