using tpa for bayesian inference

using tpa for bayesian inference

ID:7322768

大小:356.59 KB

頁數(shù):31頁

時(shí)間:2018-02-11

using tpa for bayesian inference_第1頁
using tpa for bayesian inference_第2頁
using tpa for bayesian inference_第3頁
using tpa for bayesian inference_第4頁
using tpa for bayesian inference_第5頁
資源描述:

《using tpa for bayesian inference》由會(huì)員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在工程資料-天天文庫

1、UsingTPAforBayesianInference*UniversityPressScholarshipOnlineOxfordScholarshipOnlineBayesianStatistics9JoséM.Bernardo,M.J.Bayarri,JamesO.Berger,A.P.Dawid,DavidHeckerman,AdrianF.M.Smith,andMikeWestPrintpublicationdate:2011PrintISBN-13:9780199694587PublishedtoOxfordScholarshipOnline:January2012DOI

2、:10.1093/acprof:oso/9780199694587.001.0001UsingTPAforBayesianInference*MarkHuberSarahSchottDOI:10.1093/acprof:oso/9780199694587.003.0009AbstractandKeywordsFindingtheintegratedlikelihoodofamodelgiventhedatarequirestheintegrationofanonnegativefunctionovertheparameterspace.ClassicalMonteCarlomethod

3、sfornumericalintegrationrequireaboundorestimateofthevarianceinordertodeterminethequalityoftheoutput.Themethodcalledtheproductestimatordoesnotrequireknowledgeofthevarianceinordertoproducearesultofguaranteedquality,butrequiresacoolingschedulethatmusthavecertainstrictproperties.Findingacoolingsched

4、ulecanbedifficult,andfindinganoptimalcoolingscheduleisusuallycomputationallyoutofreach.TPAisamethodthatsolvesthisdifficulty,creatinganoptimalcoolingscheduleautomaticallyasitisrun.Thismethodhasitsownsetofrequirements;hereitisshownhowtomeettheserequirementsforproblemsarisinginBayesianinference.Thi

5、sgivesguaranteedaccuracyforintegratedlikelihoodsandPage1of31UsingTPAforBayesianInference*posteriormeansofnonnegativeparameters.Keywords:AdaptiveMonteCarlo,VarianceFreeApproximationSummaryFindingtheintegratedlikelihoodofamodelgiventhedatarequirestheintegrationofanonnegativefunctionovertheparamete

6、rspace.ClassicalMonteCarlomethodsfornumericalintegrationrequireaboundorestimateofthevarianceinordertodeterminethequalityoftheoutput.Themethodcalledtheproductestimatordoesnotrequireknowledgeofthevarianceinordertoproducearesultofguaranteedquality,butrequiresacoolingschedulethatmusthavecertainstric

7、tproperties.Findingacoolingschedulecanbedifficult,andfindinganoptimalcoolingscheduleisusuallycomputationallyoutofreach.TPAisamethodthatsolvesthisdifficulty,creatinganoptimalcoolingscheduleautomaticallyasitisrun.Thismethodhas

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(dòng)畫的文件,查看預(yù)覽時(shí)可能會(huì)顯示錯(cuò)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(shí)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(shí)可能由于網(wǎng)絡(luò)波動(dòng)等原因無法下載或下載錯(cuò)誤,付費(fèi)完成后未能成功下載的用戶請聯(lián)系客服處理。