ESTIMATION AND TESTING IN THE RANDOM EFFECTS PROBIT MODEL

This paper examines the finite-sample properties of the random effects probit estimator in comparison to the standard probit estimator and the standard probit estimator with a corrected asymptotic covariance matrix. The Monte Carlo experiment considers data-generating processes consistent with longitudinal data and also data from sample surveys. The probit estimator with corrected asymptotic covariance matrix works surprisingly well over a wide range of parametric configurations and is recommended as long as an estimate of the error correlation is not of high importance.