EconStor >
Ludwig-Maximilians-Universität München (LMU) >
Sonderforschungsbereich 386: Statistische Analyse diskreter Strukturen, Universität München (LMU) >
Discussion papers, SFB 386, LMU München >

Please use this identifier to cite or link to this item:

Full metadata record

DC FieldValueLanguage
dc.contributor.authorTutz, Gerharden_US
dc.contributor.authorUlbricht, Janen_US
dc.description.abstractA new regularization method for regression models is proposed. The criterion to be minimized contains a penalty term which explicitly links strength of penalization to the correlation between predictors. As the elastic net, the method encourages a grouping effect where strongly correlated predictors tend to be in or out of the model together. A boosted version of the penalized estimator, which is based on a new boosting method, allows to select variables. Real world data and simulations show that the method compares well to competing regularization techniques. In settings where the number of predictors is smaller than the number of observations it frequently performs better than competitors, in high dimensional settings prediction measures favor the elastic net while accuracy of estimation and stability of variable selection favors the newly proposed method.en_US
dc.publisherTechn. Univ.; Sonderforschungsbereich 386, Statistische Analyse Diskreter Strukturen Münchenen_US
dc.relation.ispartofseriesDiscussion paper // Sonderforschungsbereich 386 der Ludwig-Maximilians-Universität München 486en_US
dc.subject.keywordCorrelation based estimatoren_US
dc.subject.keywordVariable selectionen_US
dc.subject.keywordElastic neten_US
dc.titlePenalized regression with correlation based penaltyen_US
dc.typeWorking Paperen_US
Appears in Collections:Discussion papers, SFB 386, LMU München

Files in This Item:
File Description SizeFormat
518770192.PDF366.99 kBAdobe PDF
No. of Downloads: Counter Stats
Show simple item record
Download bibliographical data as: BibTeX

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.