Small sample properties of a ridge regression estimator when there exist omitted variables

摘要

In this paper, we derive the exact formulae for moments of the ridge regression estimator proposed by Huang (Econ Lett 62:261–264, 1999), when there exist omitted variables.We show the conditions under which the ridge regression estimator has smallermean squared error (MSE) than the ordinary least squares estimator. Based on the exact formulae formoments, we compare the bias andMSEperformances of both estimators by numerical evaluations.