Please use this identifier to cite or link to this item:
Title: RiskLogitboost Regression for Rare Events in Binary Response: An Econometric Approach
Author: Pesantez-Narvaez, Jessica
Guillén, Montserrat
Alcañiz, Manuela
Keywords: Anàlisi de regressió
Teoria de l'estimació
Processament de dades
Sistema binari (Matemàtica)
Regression analysis
Estimation theory
Data processing
Binary system (Mathematics)
Issue Date: Mar-2021
Publisher: MDPI
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.
Note: Reproducció del document publicat a:
It is part of: Mathematics, 2021, vol. 9, num. 579, p. 1-21
Related resource:
ISSN: 2227-7390
Appears in Collections:Articles publicats en revistes (Econometria, Estadística i Economia Aplicada)

Files in This Item:
File Description SizeFormat 
710731.pdf4.43 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons