Xavier D'Haultfoeuille, CREST - ENSAE, France

Identification and estimation of average marginal effcts in fixed effect Logit models. (joint w/Laurent Davezies (CREST) & Louise Laage (Georgetown University).

No paper available.

Abstract: This article considers average marginal effects (AME) and similar parameters in a panel data fixed effects logit model. Relating the identified set of the AME to an extremal moment problem, we first show how to obtain sharp bounds on the AME straightforwardly, without any optimization. We also bound the length of the identification interval. Then, we consider two strategies to build confidence intervals on the AME. In the first, we estimate the sharp bounds with a semiparametric two-step estimator involving a first-step nonparametric estimator. We derive the asymptotic distributions of these bounds under, mainly, a restriction on the size of the support point of the unobserved heterogeneity. We consider a second very simple strategy that bypasses the nonparametric estimation but may result in larger confidence intervals. Monte Carlo simulations suggest that both approaches work well in practice, and for usual sample sizes the second is actually very competitive.​