K-L Estimator: Dealing with Multicollinearity in the Logistic Regression Model

Multicollinearity negatively affects the efficiency of maximum likelihood estimator (MLE) in both linear and generalized models. The Kibria Lukman (KLE) was developed as an alternative to MLE handle multicollinearity for regression model. In this study, we proposed Logistic Kibria-Lukman (LKLE) logistic We theoretically established superiority condition new over MLE, ridge (LRE), Liu (LLE), Liu-type (LLTE) two-parameter (LTPE) using mean squared error criteria. theoretical conditions were validated a real-life dataset, results showed that satisfied. Finally, simulation outperformed other considered estimators. However, performance estimators contingent on adopted shrinkage parameter