Robust Goodness of Fit Test Based on the Forward Search

Abstract

The most frequency used goodness of fit tests are based on measuring the distance between the theoretical distribution function and the empirical distribution function (EDF), but presence of outliers influences these tests strongly. In this study, we propose a simple robust method for goodness of fit test by using the Forward Search (FS) method. The FS method is a powerful general method for identifying outliers and their effects on the hypothesized model. The performance and the ability of the procedure to capture the structure of data, even in the presence of outliers, are illustrated by some simulation studies and real data examples.