Finite Sample Properties of FGLS Estimator for Random-Effects Model under Non-normality

Abstract

This paper considers the finite sample properties of the feasible generalized least square (FGLS) estimator for the random-effects model with non-normal errors. By using the asymptotic expansion, we study the effects of skewness and excess kurtosis on the bias and Mean Square Error (MSE) of the estimator. The numerical evaluation of our results is also presented.

Publication
Contributions to Economic Analysis, vol. 274, pp. 67-89
Date