版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請(qǐng)進(jìn)行舉報(bào)或認(rèn)領(lǐng)
文檔簡介
FinancialRiskManagementHaibinXieSchoolofBankingandFinance,UniversityofInternationalBusinessandEconomicsOffice:Boxue708TelxtremeValueTheoryEVTandVaR1BaselRulesforBacktesting2ExtremeValueTheoryandVaRBaselRulesforBacktestingTheBaselCommitteeputinplaceaframeworkbasedonthedailybacktestingofVaR.Havinguptofourexceptionsisacceptable,whichdefinesagreenzone.Ifthenumberofexceptionsisfiveormore,thebankfallsintoayelloworredzoneandincursaprogressivepenalty,whichisenforcedwithahighercapitalcharge.Roughly,thecapitalchargeisexpressedasamultiplierofthe10-dayVaRatthe99%levelofconfidence.Thenormalmultiplierkis3.Afteranincursionintotheyellowzone,themultiplicativefactor,k,isincreasedfrom3to4,orplusfactordescribedintheTableinthenextslideTheBaselPenaltyZonesAppendix1WhynormalmultiplierK=3ByChebyshevinequality:P(|x-μ|>λσ)≤1/λ2.Supposesymmetricdistribution,wegetP(x-μ<-λσ)≤1/2λ2,whichdeterminestheMaxofVaR,VaRmx=λσ.Lettheconfidencelevelbe0.99,weget1/2λ2=0.01,fromwhich,wegetλ=7.071.SupposetheusualVaRiscalculatedundertheassumptionofnormaldistribution,wegetVaRN=2.326σ.Thus,weneedamultiplierifnormaldistributionisnotsatisfied.Themultiplier,K=λσ/2.36σ=3.03Appendix2VaRParameters:TomeasuretheVaR,wefirstneedtodefinetwoquantitativeparameters:theconfidencelevelandthehorizonConfidenceLevel:Thehighertheconfidencelevel,thegreatertheVaRmeasure!Itisnotclear,however,atwhatconfidencelevelshouldonestopHorizon:Thelongerthehorizon,thegreatertheVaRmeasure.Itisnotclear,however,atwhathorizonshouldonestop.VaRParameters:SomerulesforconfidencelevelandhorizonselectionThechoiceoftheconfidencelevelandhorizondependontheintendedusefortheriskmeasures.Forbacktestingpurposes,alowconfidencelevelandashorthorizonisnecessary;forcapitaladequacypurposes,ahighconfidencelevelandalonghorizonarerequired.Inpractice,theseconflictingobjectivescanbeaccommodatedbyacomplexrule,asisthecasefortheBaselmarketriskchargeExtremeValueTheoryVaRisallaboutthetailbehavioroflossdistribution,A.K.A,weareonlyinterestedinsomeextremevalueofadistribution.D.V.GnedenkoandEVT7Бори?сВлади?мировичГнеде?нко;January1,1912–December27,1995GeneralizedParetoDistributionThishastwoparametersx(theshapeparameter)andb(thescaleparameter)Bydefinition,weexpectbtobepositive.ThecumulativedistributionisGeneralizedParetoDistributionWhenunderlingdistributionofvisnormal,wehave.increasesasthetailofvgetsheavierFormostfinancialdata,in[0.1,0.4]Thek-thmomentofunderlingr.v.isfiniteifMaximumLikelihoodEstimatorTheobservations,xi,aresortedindescendingorder.SupposethattherearenuobservationsgreaterthanuWechoosexandbtomaximizeMaximumLikelihoodEstimatorConstraintsxandbaresupposedtobepositive,althoughxnotrequiredtobepositivebythedefinitionofGPD.Negativexindicates:LightertailoftheunderlingdistributioncomparedwithnormalInappropriatevalueofuischosenFromparameterstotailofvBydefinition:ThereforeAgainsemi-parametricWhypowerlaw?ExtremeValueTheory——VaR
ExpectedShortFallBlockMaximaModelsDistributionofthelargestvariableAsngoestoinfinity,andthesupportofris[-inf,inf]WeneedtoblowupthevariablewithanormalizationThelimitingdistributionisGeneralizedExtremeValueDistributionBlockMaximaModelsGeneralizedExtremeValueDistributionVaRunderGEVdistributionAnythingwrong?BlockMaximaModelsisthedistributionofthelargestvariablenotthevariableitself.The(1-q)thquantileofrisequivalentto(1-q)^nthquantileofr(n)ThecorrectVaRis18BlockMaximaModelsEstimationBydefinitionofF*,weonlyhaveONEobservationtoestimatethreeparametersWay-outApplyGEVdistributiontomaximumreturnswithineachblockMLESelectionofnGEVisalimitproperty,naslargeaspossibleForgivenT,g=T/nwheregistheeffectivenumberofobservationsforparameterestimationBalance19MultipleperiodVaRUnderEVTthemultipleperiodVaRisnotjustsquarerootoftimehorizon.Whysquarerootoftimehorizon?Underpowerlaw Fellershowsthattailriskisapproximatelyadditive,therefore:Itiseasytoseethat 20CoherentRiskMeasures1Monotonicity:ifX1<X2,2Translationinvariance:3Homogeneity:4Subadditivity:ExerciseBasedona90%confidencelevel,howmanyexceptionsinbacktestingaVaRwouldbeexpectedovera250-daytradingyear?a.10b.15c.25d.50Alarge,internationalbankhasatradingbookwhosesizedependsontheopportunitiesperceivedbyitstraders.Themarketriskmanagerestimatestheone-dayVaR,atthe95%confidencelevel,tobe$50million.Youareaskedtobeevaluatehowgoodajobthemanagerisdoinginestimatingtheone-dayVaR.Whichofthefollowingwouldbethemostconvincingevidencethatthemanagerisdoingapoorjob,assumingthatthelossesareidenticalandindependentlydistributed(i.i.d)?a.Overthepast250days,thereareeightexceptionsb.Overthepast250days,thelargestlossis$500millionc.Overthepast250days,themeanlossis$60milliond.Overthepast250days,thereisnoexceptionWhichofthefollowingproceduresisessentialinvalidatingtheVaRestimates?a.stress-testingb.scenarioanalysisc.backtestingd.Onceapprovedbyregulators,nofurthervalidationisrequiredTheMarketRiskAmendmenttotheBaselCapitalAccorddefinestheyellowzoneasthefollowingrangeofexceptionsoutof250observationsa.3to7b.5to9c.6to9d.6to10Extremevaluetheoryprovidesvaluableinsightaboutthetailsofreturndistributions.WhichofthefollowingstatementsaboutEVTanditsapplicationsisincorrect?a.Thepeaksoverthreshold,whichthendeterminesthenumberofobservedexceedances;thethresholdmustbesufficientlyhightoapplythetheory,butsufficientlylowsothatthenumberofobservedexceedancesisareliableestimate.b.EVThighlightsthatdistributionsjustifiedbycentrallimittheoremcanbeusedforextremevalueestimationc.EVTestimatesaresubjecttoconsiderablemodelrisk,andEVTresultsareofenverysensitivetothepreciseassumptionsmaded.Becauseobserveddatainthetailsofdistributionislimited,EVestimatescanbeverysensitivetosmallsampleeffectsandotherbiasesWhichofthefollowingstatementsregardingextremevaluetheoryisincorrect?a.IncontrasttoconventionalapproachesforestimatingVaR,EVTconsidersonlythetailbehaviorofthedistributionb.ConversationalapproachesforestimatingVaRthatassumethatthedistributionofretur
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請(qǐng)下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請(qǐng)聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會(huì)有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲(chǔ)空間,僅對(duì)用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對(duì)用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對(duì)任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請(qǐng)與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時(shí)也不承擔(dān)用戶因使用這些下載資源對(duì)自己和他人造成任何形式的傷害或損失。
最新文檔
- 健康飲食文化在家庭教育中的傳播與影響研究
- 2025簡單采購合同范本
- 關(guān)于孩子在競(jìng)爭環(huán)境下自信心建立的研究
- 2025專業(yè)版委托物業(yè)管理合同樣本
- 企業(yè)培訓(xùn)提升銷售人員的客戶心理洞察力
- 以實(shí)踐為引領(lǐng)的小學(xué)科創(chuàng)教育創(chuàng)新路徑
- 2024年丹參市場(chǎng)調(diào)查報(bào)告
- 2025酒店裝修常用合同模板
- 齊魯咨詢|2024年中國新春禮盒消費(fèi)者行為洞察報(bào)告
- 2025購銷合同必須具備的合同條款
- 智慧公路交通講座-日本的智能交通與智慧公路
- 2023-2024學(xué)年教科版六年級(jí)上冊(cè)科學(xué)知識(shí)點(diǎn)總結(jié)
- 2024年甘肅定西渭源縣糧食和物資儲(chǔ)備中心選調(diào)2人歷年(高頻重點(diǎn)復(fù)習(xí)提升訓(xùn)練)共500題附帶答案詳解
- 2024年6月浙江省高考地理試卷真題(含答案)
- 2024年越南分布式光伏發(fā)電行業(yè)現(xiàn)狀及前景分析2024-2030
- 高一物理運(yùn)動(dòng)學(xué)經(jīng)典例題
- 傷口造口護(hù)理質(zhì)量標(biāo)準(zhǔn)
- Office辦公軟件理論知識(shí)考核試卷
- 客戶關(guān)系管理-課后練習(xí)參考答案 蘇朝暉
- JGJT334-2014 建筑設(shè)備監(jiān)控系統(tǒng)工程技術(shù)規(guī)范
- 可持續(xù)金融智慧樹知到期末考試答案章節(jié)答案2024年南昌大學(xué)
評(píng)論
0/150
提交評(píng)論