1、重慶大學(xué)碩士學(xué)位論文基于厚尾分布和GARCH類模型的金融風(fēng)險(xiǎn)度量研究姓名：宋鵬燕申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：概率論與數(shù)理統(tǒng)計(jì)指導(dǎo)教師：劉瓊蓀20090520重慶大學(xué)碩士學(xué)位論文英文摘要ABSTRACTAlong、Ⅳimthefinancialorgancombinationtransactionpropertynumberincreasesharplythefinancialmarketpresentstheunprecedentedund

2、ulationandthevulnerabilityT11emarketriskmeas哪eisthefoundationofcontrollingandmanagingriskAsameasurememtooloffinancialriskVaRhasbeenrecognizedbyfinancialfieldInordertomeasuretheriskaccuratelythestatisticalcharacteristicof

3、distributionmustbedescribedaboardlyFinancialAssetpricevolatilityespeciallythelargechangeforstockpricesuchasriseanddropwildlyinviewofprobabilitydistribution，studiesonoccurrencelawsforextremereturnsistoexplorethetailbehavi

4、orofreturnsprobabilitydistributionFinancialpracticalanalysisshowsthatnotonlyhastheassetreturnsfattails，butalsothevolatilityisclustered，ie1argevolatilityisfollowedbylargeonesandsmallvolatilityisfollowedbysmallvolatilityEx

5、tremevaluetheoryandmethodsCalldeal謝t11thetailbehaviorforreturnsseries，andfinancialvolatilitymodelslikeGARCHmodeldescribethereturnsvolatilitythispaperstudiesthetailbehaviorofreturnsfromthereturnsvolatilitymodels，combining

6、heavytaileddistributionwithnon—lineartimeseriesmodels，anddiscussestailfeaturesforthemarginaldistributionbythetimeseriescharacteristicnepapermakesShangzhengindexreturndams嬲researchobjects，residualsseriesareobtainedbyfitti

7、ngreturnsvolatilitymodelsinordertocomparethetailsbehaviorofinnovationsandreturnsPracticalanalysisshowsthatthemodelingmethodisabletofitthereturnstailsbetterHeavy—taileddistributionplaysallimportantroleinestimationofriskof

8、fmancialassetreturnseriesInStatisticsofExtreme，thetailindexisthebasicparameterofextremeeventsSuchaparameterplaysarelevantroleinotherextremeevents’parameters，likehighquantiles，namelyVaRinfinancialriskSoexactnessofVaRestim

9、ateddependsstronglyonexactnessofthetailindexestimatedThechapter3discussesestimatemethodsofthetailindexindetailTheclassicalHillestimatorisabiasedestimatorinmanyimprovedestimators，themaximumlikelihoodestimatorinsecondorder

10、regularconditionnotonlyisasymptoticallyunbiasedandasymptoticallynormal，butalsopracticalanalysisshowsthatthesimulationeffectiswell刪spaperdeduces1eastsquareestimatoroftailindexincommoncase(po)andprovesthattheestimatorisasy