Econometric Applications of High-Breakdown Robust Regression Techniques

Abstract

A literature search shows that robust regression techniques are rarely used in applied econometrics. We present a technique based on Rousseeuw and Van Zomeren (Journal of the American Statistical Association, 85 (1990) 633–639) that removes many of the difficulties in applying such techniques to economic data. We demonstrate the value of these techniques by re-analyzing three OLS-based regressions from the literature.

Keywords: High breakdown estimates, Masking, Robust regression, Outlier, Leverage point, Least trimmed squares (LTS), Minimum covariance determinant (MCD)

JEL Classification: C2, C52