基于小波的NN-GARCH模型及其應(yīng)用研究.pdf

1、獨(dú)創(chuàng)性聲明本人所呈交的學(xué)位論文是在導(dǎo)師指導(dǎo)下進(jìn)行的研究I作及取得的成果。盡我所知，除特別加以標(biāo)注的地方外，論文中不包含其他人的研究成果。與我一同工作的同志對本文的研究工作和成果的任何貢獻(xiàn)均已在論文中作了明確的說明并已致謝。本論文及其相關(guān)資料若有不實(shí)之處，由本人承擔(dān)一切相關(guān)責(zé)任。論文作者簽名逍訓(xùn)年歲月瑚日學(xué)位論文使用授權(quán)本人作為學(xué)位論文作者了解并愿意遵守學(xué)校有關(guān)保留、使用學(xué)位論文的規(guī)定，即：在導(dǎo)師的指導(dǎo)下創(chuàng)作完成的學(xué)位論文的知識產(chǎn)權(quán)歸西安

2、理工大學(xué)所有，本人今后在使用或發(fā)表該論文涉及的研究內(nèi)容時，會注明西安理工大學(xué)。西安理工大學(xué)擁有學(xué)位論文的如下使用權(quán)，包括：學(xué)?？梢员４鎸W(xué)位論文；可以采用影印、縮印或其他復(fù)制手段保存論文；可以查閱或借閱。本人授權(quán)西安理工大學(xué)對學(xué)位論文全部內(nèi)容編入公開的數(shù)據(jù)庫進(jìn)行檢索。本學(xué)位論文全部或部分內(nèi)容的公布(包括刊登)授權(quán)西安理工大學(xué)研究生學(xué)院辦理。經(jīng)過學(xué)校保密辦公室確定密級的涉密學(xué)位論文，按照相關(guān)保密規(guī)定執(zhí)行；需要進(jìn)行技術(shù)保密的學(xué)位論文，按照《西

3、安理工大學(xué)學(xué)位論文技術(shù)保密申請表》內(nèi)容進(jìn)行保密(附《西安理工大學(xué)學(xué)位論文技術(shù)保密申請表》)。保密的學(xué)位論文在解密后，適用本授權(quán)。做作者簽名』登翩魏莖垂i塑訓(xùn)年弓月以日AbstractTitle：THENNGARCHMODELBASEDONⅥfAVELETANDITSAPPLICATlONMajor：AppliedMathematicsName：RuoxingWANGSupervisor：ProfDeshengZHANGAbstractS

4、ignature：＆竺∑!垃巡吣qSignature：一ThisthesismainlystudiedtheNNGARCHmodelbasedonwavelettheoryanditsapplicationinpredictionofTheShanghaiStockIndexThemodelcombineswavelettheoryandneuralnetworkinfinancialtimeseriessimultaneouslybo

5、ththewaveletbasedenoisingeffectsandthestrongfittingnonlinearfunctionofneuralnetwork，alsoconsiderthefinancialtimeseriesheteroskedasticitycharacteristics，thengetsmoreaccuratefittingforecastmodelThecontentsofthethesisareasf

6、ollows：1Thewaveletanalysisintimesequenceofapplicationwerediscussed，duetothewaveletanalysishasthenatureofgoodtime—frequencydomainpropertiesand”adaptive”and”mathematicsmicroscope”，consequentlybecomecommonfocus，alsoitbecome

7、saveryusefulmathematicaltoolstotheanalysisofstockmarketSecondlydiscussedtheapplicationofneuralnetworkintimesequence，becausethestrongfittingfunctionoftheneuralnetworkhasplasticityselforganizing，adaptandverypowerfultononli

8、nearSOitalsohaswideapplicationinthefinancialtimeseriesanalysis2FirststudiedthemodelingmethodandmodelingprocessofARNNmodelbasedonwaveletanalysisHereusingbottom—upmodelingmethods，beginningthesimplestARmodel，againconsummate

9、sthemodelunceasingly；Then，appliedthemodeltotheShanghaiIndexofpredictionresearch；FinallycomparedthemodelwithARNNthatdidnotaddwaveletanalysisandsimpleARmodelTheresultshowsthatcomparednowaveletandthesimplestARmodel，theNix／A

10、Rmodelwhichbasedonwaveletanalysishasmoreapproximatingandforecastinghigherprecision3AfterthediscussionoftheAR—NNmodelbasedonwaveletanalysis，wefurtherconsideredtheresidualheteroskedasticityofthesequence，thenestimatedtheNNG

11、ARCHmodelbasedonwaveletanalysisThemodelcombineswavelettheoryandneuralnetworkwiththeGARCHmodelFirstlydealwiththenoiseofdatabythewaveletanalysis；then，estimatedthemodelwhichcontainsnonlinearneuralnetworkparametersinmeanequa