分組定時(shí)截尾數(shù)據(jù)的Bayes分析.pdf

1、燕山大學(xué)碩士學(xué)位論文分組定時(shí)截尾數(shù)據(jù)的Bayes分析姓名：胡文杰申請(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：夏茂輝張曉冉20061101燕山大學(xué)理學(xué)碩士學(xué)位論文AbstractSinceThomasBayesputforwardBayestheoremin1736，becauseoftheoutstandingrealisticbackgroundandextensiveapplicationmanyresearchersproc

2、eedthoroughresearchtoit，andithasbeenextensNelyappliedtosocialeconomyengineeringtechnique，sociallifeandproductionpracticeetc—TheresearchobjectofthepaperistheapplicatmnofBayesstatisticsanalysis，discussthedataofonlyonefailu

3、reandzero—failureingroupedtimingterminatedexperimentThemainworksofthepaperareasfollows：Inthe＆stchapter，brienyinteroducethebackgroundandthetheoryofBayesstatistics，theselectionofpriordistributionandthedevelopementofBayesst

4、atisticsathomeandabroadAtthesametime，introducethetimingterminatedexperimentanditsimprovementindetail，andmakeasummaryofzorcfailuredataproblematlast，introducethebasicconceptoftheleastsquaresmethod，ro_akeapreparationforesti

5、matethedistributionparametersInthechaptertwo，contraposeonlyonefailureinthegroupedtimingterminateddata，applytheconvexofWeibulldistributionfunctiontomakethevaluerangeoffailureprobabilitymorepreciseness，thenreceivethehierar

6、chicalBayesianestimationoftheprobabilityandimproveontheusedmethodsTheresultsofexampleandrandomsimulationindicatethatthemethodisreasonableandfeasibleinthepaperInthechapterthree，contraposethegroupedtimingterminateddataoflo

7、garithmicdistributioILintroducetheItSedmethodsofwhichestimatetheparametersofdistributionfunctionandthereliabilityforzero—failuredata，theyareclassicalmethod，Bayesianmethod，hierarchicalBayesianmethodandsynthesisestimatemet