Logistic Regression Collaborating with AI Beam Search

Tom, Daniel (2021): Logistic Regression Collaborating with AI Beam Search.

Abstract

We systematically explore the universe of all models using AI search methods. We automate much of the data preparation and testing of each model built along the way. The result is a method and system that generate superior production ready logistic regression models, beating an industry standard consumer credit risk score, GBM and NN ML models. We also incorporate into our system a method to eliminate disparate impact used by the FRB and the FTC.

• Item Type: MPRA Paper
• Language: English
• Keywords: Modeling, Regression, Logistic, AIC, IRLS, AI, ML, NN, GBM, KS, IV, GC, Wald, X2, PSI, VIF, correlation coefficient, condition index, proportion-ofvariation, reject inference, FRB, FTC, CRA, disparate impact, BISG, SBC, ARM, Intel, GPU, GPGPU, BLAS, LAPACK, transformation, normalization
• Subjects: C - Mathematical and Quantitative Methods > C6 - Mathematical Methods ; Programming Models ; Mathematical and Simulation Modeling > C61 - Optimization Techniques ; Programming Models ; Dynamic Analysis

References

1. Report to the Congress on Credit Scoring and Its Effects on the Availability and Affordability of Credit (2007, August). Board of Governors of the Federal Reserve System. https://www.federalreserve.gov/boarddocs/rptcongress/creditscore/creditscore.pdf

2. Credit-Based Insurance Scores: Impacts On Consumers Of Automobile Insurance (2007, July). A Report to Congress by the Federal Trade Commission. https://www.ftc.gov/sites/default/files/documents/reports/credit-based-insurance-scores-impactsconsumers-automobile-insurance-report-congress-federal-trade/p044804facta_report_creditbased_insurance_scores.pdf

3. Tom, Daniel, Ph.D. (2023, January 17). Eliminating Disparate Treatment in Modeling Default of Credit Card Clients. https://doi.org/10.31219/osf.io/cfyzv