简介：TheMTD(mixturetransitiondistribution)modelbasedonWeibulldistribution(WMTDmodel)isproposedinthispaper,whichisaimedatitsparameterestimation.AnEMalgorithmforestimationisgivenandshowntoworkwellbysomesimulations.Andbootstrapmethodisusedtoobtainconfidenceregionsfortheparameters.Finally,theresultsofarealexample-predictingstockprices-showthattheWMTDmodelproposedisabletocapturethefeaturesofthedatafromthick-taileddistributionbetterthanGMTD(mixturetransitiondistribution)model.

简介：Thisarticleinvestigatesthefractionalderivativeorderidentification,thecoefficientidentification,andthesourceidentificationinthefractionaldiffusionproblems.If1<α<2,weprovetheuniquedeterminationofthefractionalderivativeorderandthediffusioncoefficientp(x)by∫t0u(0,s)ds,0

简介：

简介：ThispaperconstructsaclassofHarish-Chandramoduleswithmultiplicity＜-1ofthetwoparameterdeformationofVirasoroalgebraandprovesaclassificationtheorem.

简介：Parameteridentificationproblemisoneofessentialprobleminordertomodeleffectivelyexperimentaldatabyfractalinterpolationfunction.Inthispaper,wefirstpresentanexampletoexplainarelationshipbetweeniterationprocedureandfractalfunction.Thenwediscussconditionsthatverticalscalingfactorsmustobeyinonetypicalcase.

简介：Inthispaper,anewestimatoroftheshapeparameterinthefamilyofGammadistributionisconstructedbyusingthemomentidea,anditisprovedthatthisestimatorisstronglyconsistentandasymptoticallynormal.

简介：Robustparameterdesign(RPD)isanimportantissueinexperimentaldesigns.Ifallexperimentalrunscannotbeperformedunderhomogeneousconditions,blockingtheunitsisefective.Inthispaper,weobtainthecorrespondencerelationbetweenfractionalfactorialRPDsandtheblockingschemesforfullfactorialRPDs.Inaddition,weprovideaconstructionofoptimalblockingschemesthatmakeallmainefectsandcontrol-by-noisetwo-factorinteractionsestimable.

简介：Inthispaper,weproposeadiscrepancyrule-basedmethodtoautomaticallychoosetheregularizationparametersfortotalvariationimagerestorationproblems.Theregularizationparametersareadjusteddynamicallyineachiteration.Numericalresultsareshowntoillustratetheperformanceoftheproposedmethod.

简介：Theboundedparameterestimationproblemanditssolutionleadtomoiemeaningfulresults.Itssuperiorperformanceisduetothefactthatthenewmethodguaranteesthattheeffectoftheuncertaintieswillneverbeunnecessarilyoverestimated.Wethenconsiderhowtoupdateanddowndatetheboundedparameterestimationproblem.WhenupdatinganddowndatingofSVDareusedtothenewproblem,specialtechnologiesaretakentoavoidformingUandVexplicitly,thenincreasethealgorithmperformance.BecauseofthelinkbetweentheboundedparameterestimationandTikhonovregularizationprocedure,wepointoutthatouralgorithmscanalsobeusedtomodifyregularizationproblem.

简介：

简介：<正>Thispaperconsiderstheadmissibilityoftheestimatorsforfinitepopulationwhentheparameterspaceisrestricted.Weobtainalladmissiblelinearestimatorsofanarbitrarylinearfunctionofcharacteristicvaluesofafinitepopulationintheclassoflinearestimatorsunderthecriterionoftheexpectationofmeansouarederror.

简介：Inthispaper,weconsiderageneralformoftheincrementsforatwo-parameterWienerprocess.BoththeCsorgo-Revesz'sincrementsandaclassofthelagincrementsarethespecialcasesofthisgeneralformofincrements.OurresultsimplythetheoremthathavebeengivenbyCsorgoandRevesz(1978),andsomeoftheirconditionsareremoved.

简介：在这篇文章下列随机拦截混合效果模型将被考虑：y_(ij)=v_i+x_(ij)~τβ+ε_(ij)，i=1，...，m；j=1,2，...，n_i，在此{v_i}arei.i.d。有吝啬的α和有限变化σ_v~2的随机的效果；{ε_(ij)}是有有限变化σ_ε~的i.i.d随机错误2。这里，我们将估计α，σ_v~2，σ_ε~2，β和学习他们的大样品性质例如强壮的一致性，强壮的集中率和asymptotic规度。

简介：Abstract.Usingthemethodofundeterminedfunction,asetof12parameterrectangularp|atedementwithdoub[esetparameterandgeometrysymmetryisconstructed.TheirconsistencyerrorareO(h2),oneorderhigherthantheusua[12parameterrectangu|arp[ateelements.

简介：Theparameterembeddingmethodisappliedfornumericallysolvingtheperturbedconservativesystems.BymeansofNewtonianiteration,asimplealgorithmhasbeenconstructed.Finally,theconvergenceoftheiterationisproved.

简介：Inthispaperastochasticvolatilitymodelisconsidered.Thatis,alogpriceprocessYwhichisgivenintermsofavolatilityprocessVisstudied.ThelatterisdefinedsuchthatthelogpricepossessessomeofthepropertiesempiricallyobservedbyBarndorff-Nielsen&Jiang[6].Inthemodeltherearetwosetsofunknownparameters,onesetcorrespondingtothemarginaldistributionofVandonetoautocorrelationofV.Basedondiscretetimeobservationsofthelogpricetheauthorsdiscusshowtoestimatetheparametersappearinginthemarginaldistributionandfindtheasymptoticproperties.

简介：Fourier不可分的操作员在Fourier分析和部分微分方程起一个重要作用。在这份报纸，我们处理固定双线性并且双性人参数Fourier不可分的操作员，它被一个参数FIO的学习激发并且双线性并且双性人参数Fouriermultipliers和伪differential操作员。当他们在空间变量有紧缩的支持时，我们考虑如此的FIO。如果他们包含一个真实值的阶段(x，，)它在频率变量是联合同类的，，并且顺序零的振幅支持了离开轴和antidiagonal，我们能证明固定控制local-L2盒子。一些更强壮的固定结果也在阶段功能上在更多的限制条件下面被获得。因此，我们的结果扩大固定结果为双线性并且一个参数FIO并且双线性并且双性人参数伪differential操作员到盒子双线性并且双性人参数FIO。

简介：ThepurposeofthispaperistouseaveryrecentthreecriticalpointstheoremduetoBonannoandMaranotoestablishtheexistenceofatleastthreesolutionsforthequasilinearsecondorderdifferentialequationonacompactinterval[a,b]?R{-u’’=(λf(x,u)+g(u))h(u’),in(a,b),u(a)=u(b)=0underppropriatehypotheses.Weexhibittheexistenceofatleastthree(weak)solutionsand,andtheresultsareillustratedbyexamples.

简介：Thepaperisconcernedwithsolvingperiodicboundaryproblemofsemilinearsystems,whichwillbedifferentiablyembeddedintoanone-parameterfamilyofoperators.Thesolutionofthesystemsisthenfoundbycontinuingthesolutioncurveofoperatorhomotopy.WhentheNewton-Kantorovich'sprocedureisappliedtothecorrespondingoperatorequations,anefficientalgorithmisderived.Finally,thetheoreticalresultsareinexcellentagreementwiththenumericalexamples.

简介：由使用矩阵的向量方法，我们关于线性限制学习生长曲线模型。在矩阵损失功能和向量损失功能下面，我们在不同类的线性类上为参数的线性评估者的考虑获得必要、足够的条件。