Abstract
The author proposes original tools for improving formulas of the maximum likelihood estimation method for logistic regression, the weight of evidence formula including the information value indicator formula, and the Gini index with a view to providing the use of a continuous target variable assuming probabilistic values. The approach to the pursuance of this research consists of the application of continuous weight functions meeting certain conditions to evaluate a generalized logarithm of the likelihood function including its generalized gradient vector and generalized Hessian matrix and also the application of probability theory to generalize weight of evidence and the Gini index.
Similar content being viewed by others
References
A. A. Lobanova and A. V. Chugunova, Encyclopedia of Financial Risk Management [in Russian], Alpina Publisher, Moscow (2003).
Siddiqi Naeem, Credit Risk Scorecards: Developing and Implementing Intelligent Credit Scoring, Wiley, Hoboken (2006).
Lyn C. Thomas, David B. Edelman, and Jonathan N. Crook, Credit Scoring and Its Applications, Society for Industrial and Applied Mathematics, Philadelphia (2002).
Paul D. Allison, Logistic Regression Using the SAS® System: Theory and Application, SAS Institute Inc., Cary (1999).
E. Mays, Handbook of Credit Scoring, Fitzroy Dearborn, Chicago (2001).
O. M. Soloshenko, “Improvement of the iterative classification method for including rejected applications into credit scoring,” Science News of NTUU “KPI,” No. 5, 63–69 (2014).
O. M. Soloshenko, “A method for computing the Gini index, Kolmogorov–Smirnov statistics, and Makhalanobis distance in credit scoring using SQL language tools,” Science News of NTUU “KPI,” No. 1, 29–35 (2015).
A. N. Terent’yev and P. I. Bidyuk, “Method of probabilistic inference from learning data in Bayesian networks,” Cybernetics and Systems Analysis, 43, No. 3, 391–396 (2007).
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated from Kibernetika i Sistemnyi Analiz, No. 6, November–December, 2015, pp. 174–187.
Rights and permissions
About this article
Cite this article
Soloshenko, O.M. Generalizations of Logistic Regression, Weight of Evidence, and the Gini Index for a Continuous Target Variable Taking on Probabilistic Values. Cybern Syst Anal 51, 992–1004 (2015). https://doi.org/10.1007/s10559-015-9792-z
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10559-015-9792-z