Seminars 2009 — Abstracts

Friday, June 5

Speaker: Andrey Vasnev, Sydney

Title: Sensitivity of GLS Estimators in Random Effects Models

Abstract: This paper studies the sensitivity of random effects estimators in the one-way error component regression model. Maddala and Mount (1973) give simulation evidence that in random effects models the properties of the feasible GLS estimator are not affected by the choice of the first-step estimator Š used for the covariance matrix. Taylor (1980) gives a theoretical example of this effect. This paper provides a reason for this in terms of sensitivity. The properties of Š are transferred via an uncorrelated (and independent under normality) link, called sensitivity. The sensitivity statistic counteracts the improvement in Š. A Monte Carlo experiment illustrates the theoretical findings.