Extreme Residuals in Logistic Regression Model


Abstract in English

The main objective of this paper is to obtain the extreme residuals based on two different ways. The first is the ordinary residuals and the second is the modified Pearson residuals. These two kinds of residuals are obtained through the above mentioned four methods of estimation.

References used

Aitchison J. and Silvey R. (1957) “The generalization of probit to the case of multiple response”. Biometrika; 44, 13 1-140
Aldrich J. H and Nelson F. D. (1985) “Linear probability logit and probit models” Sage publication Inc. USA
Al-Sarraf and Young, D. H. (1985) Properties of estimators of parameters in logistic regression models. Technical Report TR 105185, Departement of Mathematics and statistics, Brunel University

Download