Robustifying and simplifying high-dimensional regression with applications to yearly stock return and telematics data  

作　　者：Malvina Marchese Maria Dolores Martinez-Miranda Jens Perch Nielsen Michael Scholz 

机构地区：[1]Bayes(formerly Cass)Business School,City,University of London,106 Bunhill Row,London EC1Y 8TZ,UK [2]Department of Statistics and Operations Research,University of Granada,Campus Fuentenueva,18071Granada,Spain [3]Department of Economics,University of Klagenfurt,UniversitatsstraBe 65-67,9020 Klagenfurt,Austria [4]JOANNEUM RESEARCH Forschungsgesellschaft mbH,LeonhardstraBe 59,8010 Graz,Austria

出　　处：《Financial Innovation》2024年第1期246-261,共16页金融创新(英文)

基　　金：financial support from Ministerio de Ciencia,Innovacion y Universidades(PID2020-116587GB-I00);financial support from Austrian National Bank(Jubilaumsfondsprojekt 18901)。

摘　　要：The availability of many variables with predictive power makes their selection in a regression context difficult.This study considers robust and understandable low-dimensional estimators as building blocks to improve overall predictive power by optimally combining these building blocks.Our new algorithm is based on generalized cross-validation and builds a predictive model step-by-step from a simple mean to more complex predictive combinations.Empirical applications to annual fnancial returns and actuarial telematics data show its usefulness in the financial and insurance industries.

关 键 词：Forecasting Non-linear prediction Stock returns Dimension reduction TELEMATICS 

分 类 号：U46[机械工程—车辆工程]