Files
Document type
ArticleVersion
Published versionPublication date
Publication license
Please use this identifier to cite or link to this item: https://hdl.handle.net/2445/176310
RiskLogitboost Regression for Rare Events in Binary Response: An Econometric Approach
Journal Title
Director/Tutor
Journal ISSN
Volume Title
Related resource
Abstract
A boosting-based machine learning algorithm is presented to model a binary response with large imbalance, i.e., a rare event. The new method (i) reduces the prediction error of the rare class, and (ii) approximates an econometric model that allows interpretability. RiskLogitboost regression includes a weighting mechanism that oversamples or undersamples observations according to their misclassification likelihood and a generalized least squares bias correction strategy to reduce the prediction error. An illustration using a real French third-party liability motor insurance data set is presented. The results show that RiskLogitboost regression improves the rate of detection of rare events compared to some boosting-based and tree-based algorithms and some existing methods designed to treat imbalanced responses.
Subject (English)
Citation
Citation
PESANTEZ-NARVAEZ, Jessica, GUILLÉN, Montserrat and ALCAÑIZ, Manuela. RiskLogitboost Regression for Rare Events in Binary Response: An Econometric Approach. Mathematics. 2021. Vol. 9, num. 579, pags. 1-21. ISSN 2227-7390. [consulted: 12 of June of 2026]. Available at: https://hdl.handle.net/2445/176310