A Double-Penalized Estimator to Combat Separation and Multicollinearity in Logistic Regression

When developing prediction models for small or sparse binary data with many highly correlated covariates, logistic regression often encounters separation multicollinearity problems, resulting serious bias and even the nonexistence of standard maximum likelihood estimates. The combination makes task more difficult, a few studies addressed simultaneously. In this paper, we propose double-penalized method called lFRE to combat in regression. combines logF-type penalty ridge penalty. results indicate that compared other methods, can not only effectively remove from predicted probabilities but also provide minimum mean squared error. Aside that, real dataset is employed test performance algorithm several existing methods. result shows has strong competitiveness them be used as an alternative solve problems.