Estimation of random-effects dynamic probit models for panel data entailsthe so-called “initial conditions problem”. We argue that the relative finite-sample performance of the two main competing solutions is driven by themagnitude of the individual unobserved heterogeneity and/or of the statedependence in the data. We investigate our conjecture by means of a com-prehensive Monte Carlo experiment and offer useful indications for the practitioner.
Dynamic panel probit: finite-sample performance of alternative random-effects estimators / Lucchetti, Riccardo; Pigini, Claudia. - STAMPA. - 426:(2018), pp. 1-7.