ARCH模型參數(shù)中位無偏估計的分布.pdf

1、四川大學(xué)博士后學(xué)位論文ARCH模型參數(shù)中位無偏估計的分布姓名：王德輝申請學(xué)位級別：博士后專業(yè)：概率論與數(shù)理統(tǒng)計指導(dǎo)教師：朱允民20030601墮型盎堂苧星疊窒壬堡塑量ARCH模型參數(shù)中位無偏估計的分布AbstractInthispaperwecontinuetheresearchofthePhDthesisinthesamestudyfieldTheresultsinthePhDthesisaregeneralizedandextend

2、edintofurthercaseWedrawthefollowingthreeconclusions：First，inthePhDthesis，weprovideamedian—unbiasedestimatoroftheparameter，usingasuccinctnon—parametermethodInthispaper，weobtaintheasymptoticdistributionoftheestimatorMoreov

3、er，forsmallsamplecondition，weapplytheEdgeworthex—pansiontheoryandobtaintheapproximateformoftheestimator’sdistributionfunctionaswellasthenumericalsimulationresultsinvariouscasesSecond，inthePhDthesis，weconsiderabouttheesti

4、mateandtestoftheparametersintheARCH(O，q)modelunderorderrestrictionInthisreport，weextendalltheresultsintoARCH(p，q)modelandfindthelimitdistributionoflikelihoodratiostatisticsBasedonquadraticprogrammingmethodwealsopresentan

5、umericalsimulationaboutMLEThird，inthispaper，weuseaentropydistancefunctionininformationthe—oryasthelossfunction，andthenproposeasymmetriclossfunctionforscaleparameterwithformasL(o，d)=；5—2ForthenormaldistributionN(O，口2)，weg

6、ivetheminimumriskequivariantestimator(MRE)andtheBayesianesti—mationof口Furthermore，theadmissibilityandinadmissibilityoftheestimatorswithform(cTd)arestudiedFinally，wediscussthequestionaboutBayesianestimatorassociatedwithit