Show simple item record

dc.contributor.authorHANSEN, Peter Reinhard
dc.contributor.authorTIMMERMANN, Allan
dc.date.accessioned2012-04-02T07:45:12Z
dc.date.available2012-04-02T07:45:12Z
dc.date.issued2012
dc.identifier.issn1725-6704
dc.identifier.urihttp://hdl.handle.net/1814/21454
dc.description.abstractOut-of-sample tests of forecast performance depend on how a given data set is split into estimation and evaluation periods, yet no guidance exists on how to choose the split point. Empirical forecast evaluation results can therefore be di cult to interpret, particularly when several values of the split point might have been considered. When the sample split is viewed as a choice variable, rather than being fixed ex ante, we show that very large size distortions can occur for conventional tests of predictive accuracy. Spurious rejections are most likely to occur with a short evaluation sample, while conversely the power of forecast evaluation tests is strongest with long out-of-sample periods. To deal with size distortions, we propose a test statistic that is robust to the effect of considering multiple sample split points. Empirical applications to predictability of stock returns and inflation demonstrate that out-of-sample forecast evaluation results can critically depend on how the sample split is determined.en
dc.format.mimetypeapplication/pdf
dc.language.isoenen
dc.relation.ispartofseriesEUI ECOen
dc.relation.ispartofseries2012/10en
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectOut-of-sample forecast evaluationen
dc.subjectdata miningen
dc.subjectrecursive estimationen
dc.subjectpredictability of stock returnsen
dc.subjectinflation forecastingen
dc.subjectC12en
dc.subjectC53en
dc.titleChoice of Sample Split in Out-of-Sample Forecast Evaluationen
dc.typeWorking Paperen
dc.neeo.contributorHANSEN|Peter Reinhard|aut|EUI70016
dc.neeo.contributorTIMMERMANN|Allan|aut|
eui.subscribe.skiptrue


Files in this item

Icon

This item appears in the following Collection(s)

Show simple item record