Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap

Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap

ID:39910747

大?。?51.25 KB

頁數(shù):44頁

時間:2019-07-14

Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap_第1頁
Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap_第2頁
Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap_第3頁
Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap_第4頁
Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap_第5頁
資源描述:

《Chain Ladder Method Bayesian Bootstrap versus Classical Bootstrap》由會員上傳分享,免費在線閱讀,更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫。

1、ChainLadderMethod:BayesianBootstrapversusClassicalBootstrapGarethW.Peters1,2MarioV.W¨uthrich3PavelV.Shevchenko21UNSWMathematicsandStatisticsDepartment,Sydney,2052,Australia;email:peterga@maths.unsw.edu.au2CSIROMathematicalandInformationSciences,LockedBag17,NorthRyde,NSW,167

2、0,Australia3ETHZurich,DepartmentofMathematics,CH-8092Zurich,SwitzerlandarXiv:1004.2548v1[q-fin.CP]15Apr2010PreprintsubmittedtoElsevierApril16,2010ChainLadderMethod:BayesianBootstrapversusClassicalBootstrapGarethW.Peters1,2MarioV.W¨uthrich3PavelV.Shevchenko2AbstractTheintent

3、ionofthispaperistoestimateaBayesiandistribution-freechainladder(DFCL)modelusingapproximateBayesiancomputation(ABC)methodology.Wedemonstratehowtoestimatequantitiesofinterestinclaimsreservingandcomparetheestimatestothoseobtainedfromclassicalandcredibilityapproaches.Inthiscont

4、ext,anovelnumericalprocedureutilisingMarkovchainMonteCarlo(MCMC),ABCandaBayesianbootstrapprocedurewasdevelopedinatrulydistribution-freesetting.TheABCmethodologyarisesbecauseweworkinadistribution-freesettinginwhichwemakenoparametricassumptions,meaningwecannotevaluatethelikel

5、ihoodpoint-wiseorinthiscasesimulatedirectlyfromthelikelihoodmodel.Theuseofabootstrapprocedureallowsustogeneratesamplesfromtheintractablelikelihoodwithouttherequirementofdistributionalassumptions,thisiscrucialtotheABCframework.Thedevelopedmethodologyisusedtoobtaintheempirica

6、ldistributionoftheDFCLmodelparametersandthepredictivedistributionoftheoutstandinglossliabilitiesconditionalontheobservedclaims.WethenestimatepredictiveBayesiancapitalestimates,theValueatRisk(VaR)andthemeansquareerrorofprediction(MSEP).Thelatteriscomparedwiththeclassicalboot

7、strapandcredibilitymethods.Keywords:Claimsreserving,distribution-freechainladder,meansquareerrorofprediction,Bayesianchainladder,approximateBayesiancomputation,MarkovchainMonteCarlo,annealing,bootstrapPreprintsubmittedtoElsevierApril16,20101.MotivationThedistribution-freech

8、ainladdermodel(DFCL)ofMack[14]isapopularmodelforstochasticclaimsreserving.Inthispa

當(dāng)前文檔最多預(yù)覽五頁,下載文檔查看全文

此文檔下載收益歸作者所有

當(dāng)前文檔最多預(yù)覽五頁,下載文檔查看全文
溫馨提示:
1. 部分包含數(shù)學(xué)公式或PPT動畫的文件,查看預(yù)覽時可能會顯示錯亂或異常,文件下載后無此問題,請放心下載。
2. 本文檔由用戶上傳,版權(quán)歸屬用戶,天天文庫負(fù)責(zé)整理代發(fā)布。如果您對本文檔版權(quán)有爭議請及時聯(lián)系客服。
3. 下載前請仔細(xì)閱讀文檔內(nèi)容,確認(rèn)文檔內(nèi)容符合您的需求后進(jìn)行下載,若出現(xiàn)內(nèi)容與標(biāo)題不符可向本站投訴處理。
4. 下載文檔時可能由于網(wǎng)絡(luò)波動等原因無法下載或下載錯誤,付費完成后未能成功下載的用戶請聯(lián)系客服處理。