A0398
Title: A bootstrap comparison of robust regression estimators
Authors: Jan Kalina - The Czech Academy of Sciences, Institute of Information Theory and Automation (Czech Republic)
Patrik Janacek - The Czech Academy of Sciences, Institute of Information Theory and Automation (Czech Republic) [presenting]
Abstract: The least squares estimator in linear regression is well known to be highly vulnerable to the presence of outliers in the data. Available robust statistical estimators are preferable as alternatives to the classical least squares. It has been repeatedly recommended to use the least squares together with a robust estimator, where the latter is understood as a diagnostic tool for the former. In other words, only if the robust estimator yields a very different result, the user should investigate the dataset closer and search for explanations. This requires a formal hypothesis test. A bootstrap test of equality of two linear regression estimators is developed. Its performance is presented on several real economic datasets contaminated by outliers. Although robust estimation (and particularly the least weighted squares estimator) is beneficial in all these datasets, robust estimates turn out not to be significantly different from non-robust ones.
