V-mask Type Criterion for Identification of Outliers in Logistic Regression

Abstract

A procedure is proposed to identify multiple outliers in the logistic regression. It detects the leverage points by means of hierarchical clustering of the robust distances based on the minimum covariance determinant estimator, and then it employs a V-mask criterion on the scatter plot of robust residuals against robust distances to classify the observations into vertical outliers, bad leverage points, good leverage points, and regular points. Efectiveness of the proposed procedure is evaluated on the basis of the clasic and artificial data sets, and it is shown that the procedure deals very well with the masking and swamping effects.