简介:时空单极方程是在到ℝ2,1的ℝ2,2的anti-self-dualYang工厂方程的减小。这个方程是一个非线性的波浪方程,并且能被编码为宽松的对。相等的宽松的对被奶妈和Terng使用与连续散布数据构造单极,然后方程能被散布数据线性化,允许一个使用散布方法解决有很快腐烂的Cauchy问题的逆小起始的数据。在这篇论文,我们使用holomorphic捆的术语和某些地图的transversality,由起始的数据的parametrized,多达计量转变给更起始的数据,我们能与使用散布方法解决单极方程的Cauchy问题。
简介:Semiparametrictransformationmodelsprovideaclassofflexiblemodelsforregressionanalysisoffailuretimedata.Severalauthorshavediscussedthemunderdifferentsituationswhencovariatesaretimeindependent(Chenetal.,2002;Chengetal.,1995;Fineetal.,1998).Inthispaper,weconsiderfittingthesemodelstoright-censoreddatawhencovariatesaretime-dependentlongitudinalvariablesand,furthermore,maysuffermeasurementerrors.Forestimation,weinvestigatethemaximumlikelihoodapproach,andanEMalgorithmisdeveloped.Simulationresultsshowthattheproposedmethodisappropriateforpracticalapplication,andanillustrativeexampleisprovided.
简介:Itisknownthatconditionalindependenceisaquitebasicassumptioninmanyfieldsofstatistics.Howtotestitsvalidityisofgreatimportanceandhasbeenextensivelystudiedbytheliterature.Nevertheless,alloftheexistingmethodsfocusonthecasethatdataarefullyobserved,butnoneofthemseemshavingtakenintoaccountofthescenariowhenmissingdataarepresent.Motivatedbythis,thispaperdevelopstwotestingstatisticstohandlesuchasituationrelyingontheideaofinverseprobabilityweightedandaugmentedinverseprobabilityweightedtechniques.Theasymptoticdistributionsoftheproposedstatisticsarealsoderivedunderthenullhypothesis.Thesimulationstudiesindicatethatbothtestingstatisticsperformwellintermsofsizeandpower.
简介:Receiveroperatingcharacteristic(ROC)curvesareoftenusedtostudythetwosampleprobleminmedicalstudies.However,mostdatainmedicalstudiesarecensored.UsuallyanaturalestimatorisbasedontheKaplan-Meierestimator.InthispaperweproposeasmoothedestimatorbasedonkerneltechniquesfortheROCcurvewithcensoreddata.Thelargesamplepropertiesofthesmoothedestimatorareestablished.Moreover,deficiencyisconsideredinordertocomparetheproposedsmoothedestimatoroftheROCcurvewiththeempiricalonebasedonKaplan-Meierestimator.ItisshownthatthesmoothedestimatoroutperformsthedirectempiricalestimatorbasedontheKaplan-Meierestimatorunderthecriterionofdeficiency.Asimulationstudyisalsoconductedandarealdataisanalyzed.
简介:Thepurposeofthepresentpaperistocallforattentiontothefollowingquestion:Whichoftheinitialdata(nonsmall)admitglobalsmoothsolutionstotheCauchyproblemfornonlinearwaveequations.Afewcasesandexamplesaresketched,showingthatthegeneralanswerofthisquestionmaybequitecomplicated.
简介:Theboundedparameterestimationproblemanditssolutionleadtomoiemeaningfulresults.Itssuperiorperformanceisduetothefactthatthenewmethodguaranteesthattheeffectoftheuncertaintieswillneverbeunnecessarilyoverestimated.Wethenconsiderhowtoupdateanddowndatetheboundedparameterestimationproblem.WhenupdatinganddowndatingofSVDareusedtothenewproblem,specialtechnologiesaretakentoavoidformingUandVexplicitly,thenincreasethealgorithmperformance.BecauseofthelinkbetweentheboundedparameterestimationandTikhonovregularizationprocedure,wepointoutthatouralgorithmscanalsobeusedtomodifyregularizationproblem.
简介:GeometricprocesswasfirstintroducedbyLam^[10,11].Astochasticprocess{Xi,i=1,2,...}iscalledageometricprocess(GP)if,forsomeα>0,{a^i-1Xi,i=1,2,...}formsarenewalprocess.Inthispaper,theGPisusedtoanalyzethedatafromaseriesofevents.AnonparametricmethodisintroducedfortheestimationofthethreeparametersintheGP.Thelimitingdistributionsofthethreeestimatorsarestudied.Throughtheanalysisofsomerealdatasets,theGPmodeliscomparedwithotherthreehomogeneousandnonhomogeneousPoissonmodels.ItseemsthatonaveragetheGPmodelisthebestmodelamongthesefourmodelsinanalyzingthedatafromaseriesofevents.
简介:在这篇论文,我们与间隔参数为网络连接问题介绍一个得最高分的战略模型。我们考虑怎么在一项给定的预算下面与一条路径或一棵跨越的树在一个网络连接给定的节点,在每个连接与间隔被联系并且能在间隔以任何值的成本被建立的地方。一个单个连接的质量(或连接失败的风险,等等)取决于它的构造费用和联系间隔。完成网络连接的公平,我们在在所有连接上的最大的风险的最小化的模型目的使用了。我们建议分别地在多项式时间发现最佳的路径和跨越的树的二个算法。多项式解决之可能性与间隔数据为网络连接显示我们的得最高分的战略模型和柔韧的偏差标准的模型之间的突出的差别,它产生NP难的优化问题。
简介:Recurrenteventdataoftenarisesinbiomedicalstudies,andindividualswithinaclustermightnotbeindependent.Weproposeasemiparametricadditiveratesmodelforclusteredrecurrenteventdata,whereinthecovariatesareassumedtoaddtotheunspecifiedbaselinerate.Fortheinferenceonthemodelparameters,estimatingequationapproachesaredeveloped,andbothlargeandfinitesamplepropertiesoftheproposedestimatorsareestablished.
简介:ThispapergivesadefinitionofpermanentoptimaldatapointofLeastAbsoluteDeviation(LAD)problem.Sometheoreticalresultsonnon-degenerateLADproblemareobtained.ForcomputingLADproblem,anefficient,algorithmisgivenaccordingtotheideaofpermanentoptimaldatapoint.Numericalexperienceshowsthatouralgorithmisbetterthanmanyofothers,includingthefamousBRalgorithm.