简介：ThegeneralresultsonconvergenceoftheIshikawaiterationprocedureswitherrorsforLipschitzianφ-strongpseudo-contractionsandnonlinearoperatorequationsofφ-stronglyaccretivetypeisestablishedinarbitraryBanachspaces.Asthedirectapplications,somestabilityresultsoftheIshikawaiterationmethodsforψ-strongpseudo-contractionsandnonlinearoperatorequationsofφstrnglyaccretivetypearealsogiven.OurresultsinthispaperimproveandextendtherecentresultsduetoOsilikeandotherauthors.

简介：Inordertogiveacompleteandaccuratedescriptionaboutthesensitivityofefficientportfoliostochangesinassets'expectedreturns,variancesandcovariances,thejointeffectofestimationerrorsinmeans,variancesandcovariancesontheefficientportfolio'sweightsisinvestigatedinthispaper.Itisprovedthattheefficientportfolio'scompositionisaLipschitzcontinuous,differentiablemappingoftheseparametersundersuitableconditions.Thechangerateoftheefficientportfolio'sweightswithrespecttovariationsaboutriskreturnestimationsisderivedbyestimatingtheLipschitzconstant.Ourgeneralquantitativeresultsshowthattheefficientportfolio''sweightsarenormallynotsosensitivetoestimationerrorsaboutmeansandvariances.Moreover,wepointoutthoseextremecaseswhichmightcausestabilityproblemsandhowtoavoidtheminpractice.Preliminarynumericalresultsarealsoprovidedasanillustrationtoourtheoreticalresults.

简介：Thispaperstudiesthemodel-robustdesignproblemforgeneralmodelswithanunknownbiasorcontaminationandthecorrelatederrors.ThetrueresponsefunctionisassumedtobefromareproducingkernelHilbertspaceandtheerrorsarefittedbytheqthordermovingaverageprocessMA(q),especiallytheMA(1)errorsandtheMA(2)errors.Inbothsituations,designcriteriaarederivedintermsoftheaverageexpectedquadraticlossfortheleastsquaresestimationbyusingaminimaxmethod.Acaseisstudiedandtheorthogonalityofthecriteriaisprovedforthisspecialresponse.Therobustnessofthedesigncriteriaisdiscussedthroughseveralnumericalexamples.

简介：InthispaperthelimitingdistributionoftheleastsquareestimatefortheautoregressivecoefficientofanearlyunitrootmodelwithGARCHerrorsisderived.Sincethelimitingdistributiondependsontheunknownvarianceoftheerrors,anempiricallikelihoodratiostatisticisproposedfromwhichconfidenceintervalscanbeconstructedforthenearlyunitrootmodelwithoutknowingthevariance.Togainanintuitivesensefortheempiricallikelihoodratio,asmallsimulationfortheasymptoticdistributionisgiven.

简介：ThevalueofaEuropeanoptionsatisfiestheBlack-Scholesequationwithappropriatelyspecifiedfinalandboundaryconditions.Wetransformtheproblemtoaninitialboundaryvalueproblemindimensionlessform.Therearetwoparametersinthecoefficientsoftheresultinglinearparabolicpartialdifferentialequation.Forarangeofvaluesoftheseparameters,thesolutionoftheproblemhasaboundaryoraninitiallayer.Theinitialfunctionhasadiscontinuityinthefirst-orderderivative,whichleadstotheappearanceofaninteriorlayer.Weconstructanalyticallytheasymptoticsolutionoftheequationinafinitedomain.Basedontheasymptoticsolutionwecandeterminethesizeoftheartificialboundarysuchthattherequiredsolutioninafinitedomaininxandatthefinaltimeisnotaffectedbytheboundary.Also,westudycomputationallythebehaviourinthemaximumnormoftheerrorsinnumericalsolutionsincasessuchthatoneoftheparametersvariesfromfinite(orprettylarge)tosmallvalues,whiletheotherparameterisfixedandtakeseitherfinite(orprettylarge)orsmallvalues.Crank-Nicolsonexplicitandimplicitschemesusingcenteredorupwindapproximationstothederivativearestudied.Wepresentnumericalcomputations,whichdetermineexperimentallytheparameter-uniformratesofconvergence.Wenotethatthisrateisratherweak,dueprobablytomixedsourcesoferrorsuchasinitialandboundarylayersandthediscontinuityinthederivativeofthesolution.

简介：在这份报纸，我们在灯芯的意义在一个部分Black-Scholes模型调查分离时间hedging策略的错误的asymptotic行为--它?-Skorohod集成。hedging错误的集中的率由于分离时间的做贸易当真策略以商人著名时，被调查。结果提供新统计工具学习并且检测效果长记忆并且为分离时间hedging的错误的林中小丘参数。

简介：Inthispaper，theempiricallikelihoodconfidenceregionsfortheregressioncoefficientinalinearmodelareconstructedunderm-dependenterrors.Itisshownthattheblockwiseempiricallikelihoodisagoodwaytodealwithdependentsamples.

简介：在这份报纸，当错误形成一个严格地静止的混合序列时，为在一个线性模型的回归参数的M评估者的中等偏差被获得。结果被使用学习象Hubers评估者，Lp回归评估者，最少的广场评估者和最不绝对的偏差评估者那样的M评估者的许多不同类型。

简介：让Bp，1p<，是从Lp的所有围住的功能的空间()它能被扩大到指数的类型的全部功能。没有腐烂假设，为截断的Whittaker-Kotelnikov-Shannon系列的一致错误界限在无穷为功能fBp基于本地采样被导出。然后aliasing错误和为从Sobolev类U的non-bandlimited函数的Whittaker-Kotelnikov-Shannon扩大的截断错误的最佳的界限(Wpr())被决定直到一个对数的因素。

简介：Thispaperproposesanewapproachforvariableselectioninpartiallylinearerrors-in-variables(EV)modelsforlongitudinaldatabypenalizingappropriateestimatingfunctions.WeapplytheSCADpenaltytosimultaneouslyselectsignificantvariablesandestimateunknownparameters.Therateofconvergenceandtheasymptoticnormalityoftheresultingestimatorsareestablished.Furthermore,withproperchoiceofregularizationparameters,weshowthattheproposedestimatorsperformaswellastheoracleprocedure.Anewalgorithmisproposedforsolvingpenalizedestimatingequation.Theasymptoticresultsareaugmentedbyasimulationstudy.

简介：在这篇论文，我们为解决与错误建议并且分析一个三步的反复的计划非线性强烈没有Lipshitzcondition的accretive操作员方程Tx=f。在这篇论文介绍的结果在更一般的背景改进并且延长当前的结果。

简介：Let1＜ρ≤2,Ebearealρ-uniformlysmoothBanachspaceandT:E→Ebeacontinuousandstronglyaccretiveoperator.ThepurposeofthispaperistoinvestigatetheproblemofapproximatingsolutionstotheequationTx=fbytheIshikawaiterationprocedurewitherrors(?)wherex_0∈E,{u_n},{υ_n}areboundedsequencesinEand{α_n},{b_n},{c_n},{a_n~'},{b_n~'},{c_n~'}arerealsequencesin[0,1].Undertheassumptionofthecondition0＜α≤b_n+c_n,An≥0,itisshownthattheiterativesequence{x_n}convergesstronglytotheuniquesolutionoftheequationTx=f.Furthermore,undernoassumptionofthecondition(?)(b_n~'+c_n~')=0,itisalsoshownthat{x_n}convergesstronglytotheuniquesolutionofTx=f.

简介：Thepurposeofthispaperistoinvestigatesomesufficientandnecessaryconditionsforthree-stepIshikawaiterativesequenceswitherrortermsforuniformlyquasi-Lipschitzianmappingstoconvergetofixedpoints.OurresultsextendandimprovetherecentonesannouncedbyLiu[3,4],XuandNoor[5],andmanyothers.

简介：混乱理论教了我们很可能有非线性和随机的输入愿望的一个系统生产不规则的数据。如果随机的错误是不规则的数据，那么随机的错误过程将提起非线性(Kantz和Schreiber(1997))。Tsai(1986)与AR(1)错误在线性模型为自相关和heteroscedasticity介绍了合成测试。刘(2003)与DBL在非线性的模型为关联和heteroscedasticity介绍了合成测试(p，0,1)错误。因此，在回归模型的重要问题是bilinearity，关联和heteroscedasticity的察觉。在这篇文章，作者与DBL讨论非线性的模型的更一般的大小写(p，q，1)由20测试的随机的错误。为bilinearity，关联，和heteroscedasticity的测试的几统计在简单矩阵公式被获得，并且表示。有线性错误的回归模型的结果与双线性的错误被扩大到那些。模拟学习被执行调查测试统计的力量。这篇文章的所有结果扩大并且发展结果Tsai(1986)，魏，等(1995)，和刘，等(2003)。

简介：Inthispaper,weinvestigatetheproblemofapproximatingsolutionsoftheequationsofLipschitzianψ-stronglyaccretiveoperatorsandfixedpointsofLipschitzianψ-hemicontractiveoperatorsbylshikawatypeiterativesequenceswitherrors.Ourresultsunify,improveandextendtheresultsobtainedpreviouslybyseveralauthorsincludingLiandLiu(ActaMath.Sinica41(4)(1998),845-850),andOsilike(NonlinearAnal.TMA,36(1)(1999),1-9),andalsoanswercompletelytheopenproblemsmentionedbyChidume(J.Math.Anal.Appl.151(2)(1990),453-461).

简介：Semiparametrictransformationmodelsprovideaclassofflexiblemodelsforregressionanalysisoffailuretimedata.Severalauthorshavediscussedthemunderdifferentsituationswhencovariatesaretimeindependent(Chenetal.,2002;Chengetal.,1995;Fineetal.,1998).Inthispaper,weconsiderfittingthesemodelstoright-censoreddatawhencovariatesaretime-dependentlongitudinalvariablesand,furthermore,maysuffermeasurementerrors.Forestimation,weinvestigatethemaximumlikelihoodapproach,andanEMalgorithmisdeveloped.Simulationresultsshowthattheproposedmethodisappropriateforpracticalapplication,andanillustrativeexampleisprovided.

简介：Theauthorsstudytheempiricallikelihoodmethodforpartiallylinearerrors-in-variablesmodelwithcovariatedatamissingatrandom.Empiricallikelihoodratiosfortheregressioncoefficientsandthebaselinefunctionareinvestigated,andthecorrespondingempiricallog-likelihoodratiosareprovedtobeasymptoticallystandardchi-squared,whichcanbeusedtoconstructconfidenceregions.Thefinitesamplebehavioroftheproposedmethodsisevaluatedbyasimulationstudywhichindicatesthattheproposedmethodsarecomparableintermsofcoverageprobabilitiesandaveragelengthofconfidenceintervals.Finally,theEarthquakeMagnitudedatasetisusedtoillustrateourproposedmethod.