Multicollinearity problem in the logistic regression model

In logistic regression model, the most popular parameter estimation method is maximum likelihood. In the existence of multicollinearity, the length and the asymptotic covariance of the maximum likelihood estimator are excessively large. As a result, statistical inferences become imprecise. In order to reduce the effects of multicollinearity, some alternative parameter estimation methods to maximum likelihood were proposed in the literature. In this study, first-order jackknifed ridge logistic estimator, first-order r-k and r-d class logistic estimators are introduced. The properties and the performances of these estimators are examined. The theoretical results are illustrated via numerical examples and simulation studies. Key Words: Logistic regression, Multicollinearity, First-order jackknifed ridge logistic estimator, First-order r-k class logistic estimator, First-order r-d class logistic estimator