24360.基于二階預(yù)測有效度的組合預(yù)測模型研究

1、獨創(chuàng)性聲明本人聲明所呈交的學(xué)位論文是本人在導(dǎo)師指導(dǎo)下進行的研究工作及取得的研究成果。據(jù)我所知，除了文中特別加以標注和致謝的地方外，論文中不包含其他人已經(jīng)發(fā)表或撰寫過的研究成果，也不包含為獲得圣徼大學(xué)或其他教育機構(gòu)的學(xué)位或證書而使用過的材料。與我一同工作的同志對本研究所做的任何貢獻均已在論文中作了明確的說明并表示謝意。學(xué)位論文作者簽名：云3問j簽字日期：珈f爭年年月弓。日學(xué)位論文版權(quán)使用授權(quán)書本學(xué)位論文作者完全了解圣徼犬巷有關(guān)保留、使用學(xué)

2、位論文的規(guī)定，有權(quán)保留并向國家有關(guān)部門或機構(gòu)送交論文的復(fù)印件和磁盤，允許論文被查閱和借閱。本人授權(quán)喧胬木巷可以將學(xué)位論文的全部或部分內(nèi)容編入有關(guān)數(shù)據(jù)庫進行檢索，可以采用影印、縮印或掃描等復(fù)制手段保存、匯編學(xué)位論文。(保密的學(xué)位論文在解密后適用本授權(quán)書)學(xué)位論文作者簽名：l丹、導(dǎo)師簽名：形彳友簽字日期：占嘩年名月；9日學(xué)位論文作者畢業(yè)去向：工作單位：通訊地址：簽字日期：b，留年幺月；口日電話：郵編：AbstractCombinationf

3、orecastingisthe印proachthatcombinedseVemldi虢rentforecastingmetllodsthrough廿1eappropriate、vei曲tedaVeragingmethods，sothatthee毹ctiVeinfomationwllichisproVidedbyt11eV撕ouSpredictiVemethodscaIlbeutilizedcomprehensiVelyThebasicp

4、rincipleofcombinationforecastingistocombineeVeUsinglepredictiVeValueaIlddi虢rentweights血oasin91epredictiVevalue，tllenthee11rorbe帆eenforecaStiIlgandactllalwouldbereduced，somatthepredictiVeaccuIacycanbeimpr0VedmoreefrectiVe

5、Atpresent，mostoflinearcombinationforecaStingisbaSedontllepredictiVecriteriathattheminimumVariarlcemeⅡ10dorliIlearregressionmemodtoestablishButinfact，suchguidelinesandassuI】叩tionscouldnot、ⅣeUrenectthee行ectiVenessofthepred

6、ictionmethodBecausethedi疵rentindicatorsserieshaVedifferentdimensions，sothemillimumVariaIlcemethodorlinearre鏟essionmetllodalsodonothavethesamedimensionIIlttlisregard，mispapersetupthecombinationforecaStingmodelbyusingⅡleme

7、anofpredictionaccuracyaIldtheVariallce、)v：hjchcaIlrenectthediscretede鏟eeThemaincontentsare搠gedaSfollows：ChapterI：7rhecurrentdoⅡlesticaIldintemationalresearchstatusofcombinationforecaStingandmecriteriausedbycombinationfor

8、ecaStingaredescribedbrienyCll叩terII：PreliminarykrlowledgeIntllispart，theconceptandtherelatedmathematicaloperationmlesofinten，alnuIllberareillustrated，meconcepta11drelatedpropeniesofinducedorderedweightedaverageoperatorar

9、eiIl扛oducedChapterIII：AcombinationforecaStmgmodelwithgeometricmeanisestablished，whichisbasedonsecondorderfIorecaSte丘IectiVemeaSure，mentherelatedpropertiesofmemodelisdiscussedSomeconceptssuChaSsuperiorconl_binationforecas

10、ting，redundaIltforecastingaIldsoonareinnoducedFinaJly，an眥1嘶calex鋤pleisgiVentoaIlalyzetheea、ectiVenessoftlleproposedcombinationforecaStmodelChapterIV：CombinedsecondorderforecaStefl’ectiVemeasurea11dinducedorderedwei曲tedav