The application of brute force logistic regression to corporate credit scoring models: Evidence from Serbian financial statements

作者：

Highlights：

• A model for real-life credit scoring application has been developed using brute force logistic regression.

• A thorough investigation of most predictive financial ratios has been performed.

• A clustering technique for reducing the number of highly-correlated variables has been applied.

• A model with highest predictive power has been proposed.

摘要

•A model for real-life credit scoring application has been developed using brute force logistic regression.•A thorough investigation of most predictive financial ratios has been performed.•A clustering technique for reducing the number of highly-correlated variables has been applied.•A model with highest predictive power has been proposed.

论文关键词：Credit scoring,Logistic regression,Probability of default,Weight of evidence approach,Corporate entities

论文评审过程：Available online 11 May 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.05.022