nonparametric prediction in time series analysis some empirical results

nonparametric prediction in time series analysis some empirical results

ID:7289045

大小:173.73 KB

页数:10页

时间:2018-02-10

nonparametric prediction in time series analysis  some empirical results_第1页
nonparametric prediction in time series analysis  some empirical results_第2页
nonparametric prediction in time series analysis  some empirical results_第3页
nonparametric prediction in time series analysis  some empirical results_第4页
nonparametric prediction in time series analysis  some empirical results_第5页
资源描述:

《nonparametric prediction in time series analysis some empirical results》由会员上传分享,免费在线阅读,更多相关内容在工程资料-天天文库

1、Nonparametricpredictionintimeseriesanalysis:someempiricalresultsMarcellaNiglioandCiraPernaAbstract.InthispaperanewapproachtoselectthelagpfortimeseriesgeneratedfromMarkovprocessesisproposed.Itisfacedinthenonparametricdomainanditisbasedontheminimisationoftheestimatedris

2、kofpredictionofone-step-aheadkernelpredictors.TheproposedprocedurehasbeenevaluatedthroughaMonteCarlostudyandinempiricalcontexttoforecasttheweakly90-dayUST-billsecondarymarketrates.Keywords:kernelpredictor,estimatedriskofprediction,subsampling1IntroductionOneoftheaimsi

3、ntimeseriesanalysisisforecastingfuturevaluestakingadvantageofcurrentandpastknowledgeofthedata-generatingprocesses.Thesestructuresareoftensummarisedwithparametricmodelsthat,basedonspecificassumptions,definetherelationshipsamongvariables.Inthisparametriccontextalargenumbe

4、rofmodelshavebeenproposed(amongothers,[3],[20],[4],and,morerecently,[11],whichdiscussesparametricandnonparametricmethods)andformostofthemtheforecastperformancehasbeenevaluated.Toovercometheproblemofpriorknowledgeaboutthefunctionalformofthemodel,anumberofnonparametricm

5、ethodshavebeenproposedandwidelyusedinstatisticalapplications.Inthiscontext,ourattentionisfocusedonnonparametricanalysisbasedonkernelmethodswhichhavereceivedincreasingattentionduetotheirflexibilityinmodellingcomplexstructures.Inparticular,givenaMarkovprocessoforderp,int

6、hispaperanewapproachtoselectthelagpisproposed.Itisbasedontheminimisationoftheriskofprediction,proposedin[13],estimatedforkernelpredictorsbyusingthesubsampling.Afterpresentingsomeresultsonthekernelpredictors,wediscuss,inSection2,howtheycanbeintroducedintheproposedproce

7、dure.InSection3wefurtherdescribethealgorithmwhoseperformancehasbeendis-cussedinaMonteCarlostudy.Toevaluatetheforecastaccuracyofthenonparametricpredictorinthecontextofrealdata,inSection4wepresentsomeresultsontheweeklyM.Corazzaetal.(eds.),MathematicalandStatisticalMetho

8、dsforActuarialSciencesandFinance©Springer-VerlagItalia2010236MarcellaNiglioandCiraPerna90-dayUST-billsecondarymarketrates.So

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。