简介:Theobjectiveofthispaperistogiveaconstructiveapproachtothesolutionofthefundamentalfunctionsforcardinalinterpolationfromashift-invariantspacegeneratedbythe(multi-)integertranslatesofsomecompactlysupportedfunctionwhosepolynomialsymbolhasanon-emptyzeroset.ThisproblemwasfirstintroducedbyChut,Diamond,andRaphael,whereexplicitsolutionsweregivenforvariouszerosets.Later,deBoor,Hollig,andRiemenschneidergaveanexistenceproofforzerosetswhicharemoregeneral.Inthispa-per,wegiveanintegralrepresentationofthefundamentalsolutionsthatcanbemadeexplicitinsomecasesandwewillalsogiveagrowthconditionofsuchfundamentalsolutions.Thefour-directionalboxsplineswillbeusedasanillustrativeexample.
简介:在这笔记,我们学习多变量伯恩斯坦多项式B的本地行为,在d维的单一的SRd上。当n趋于到无穷,为同意足够的高顺序的衍生物进入x的功能f,我们导出Bnf的完全的asymptotic扩大。n1的所有系数仅仅取决于f和x明确地被计算。结果组合的数字起一个重要作用。同等结果以一个方法由于R.张概括最近的公式。
简介:Thispaperanalyzesthedynamicsofnonlinearmultivariatetimeseriesmodelsthatisrepresentedbygeneralizedimpulseresponsefunctionsandasymmetricfunctions.Weillustratethemeasuresofshockpersistencesandasymmetriceffectsofshocksderivedfromthegeneralizedimpulseresponsefunctionsandasymmetricfunctioninbivariatesmoothtransitionregressionmodels.TheempiricalworkinvestigatesabivariatesmoothtransitionmodelofUSGDPandtheunemploymentrate.
简介:MultivariateObjectiveAnalysisofWindandHeightFieldsintheTropicsS.K.Sinha,S.G.NarkhedkarandS.Rajamani(IndianinstituteofTropical...
简介:从截断的multivariate正常分布(TMVND)的采样组成在适合统计的许多和econometric的计算模块建模的核心。我们建议二个有效方法,一个反复的数据扩大(DA)算法和非反复的逆Bayes公式(IBF)刺绣花样,模仿TMVND并且与线性不平等限制概括他们到multivariate正常分布。由创造贝叶斯的不完全数据的结构,DA算法的以后的步直接产生随机的向量与单个元素对比绘画绘画,产生明显的计算优点并且与象S加,MATLAB和高斯那样的普通统计软件包裹的容易的编码。而且,DA为实现一个快他们算法识别TMVND的模式提供准备好了的结构,它在抑制参数问题的统计推理有许多潜在的应用程序。另外,作为中间的结果利用这个模式,IBF采样提供一种新奇选择给吉布斯采样并且由于在TMVND的部件之间的高关联与集中和可能的慢集中消除问题。DA算法与抑制参数被用于一个线性回归模型并且与一个出版数据集合被说明。数字比较证明建议DA算法和IBF刺绣花样比吉布斯刺绣花样和接受拒绝算法更有效。
简介:ThispaperpresentsanewalgorithmforcomputingtheextendedHenselconstruction(EHC)ofmultivariatepolynomialsinmainvariablexandsub-variablesu1,u2,···,umoveranumberfieldK.Thisalgorithmfirstconstructsasetbyusingtheresultantoftwoinitialcoprimefactorsw.r.t.x,andthenobtainstheHenselfactorsbycomparingthecoefficientsofxionbothsidesofanequation.SincetheHenselfactorsarepolynomialsofthemainvariablewithcoefficientsinfractionfieldK(u1,u2,···,um),thecomputationcostofhandlingrationalfunctionscanbehigh.Therefore,theauthorsuseamethodwhichmultipliesresultantandremovesthedenominatorsoftherationalfunctions.Unlikepreviously-developedalgorithmsthatuseinterpolationfunctionsorGrobnerbasis,thealgorithmrelieslittleonpolynomialdivision,andavoidsmultiplyingbydifferentfactorswhenremovingthedenominatorsofHenselfactors.AllalgorithmsareimplementedusingMagma,acomputationalalgebrasystemandexperimentsindicatethatouralgorithmismoreefficient.
简介:OjbectiveTofindtheindependentpredictorsforrestenosisaftercoronarystenting.MethodsQuantitativeangiographywasperformedon60cases(67successfullydilatedlesions)afterangio-plastyover6-monthsfollow-up,andbothunivariateandmultivariatelogisticregressionanalysisweredonetoi-dentifythecorrelationsofrestenosiswithclinicalfactors.ResultsThetotalrestenosisratewas31.3%(21of67lesions),andaccordingtounivariateanalysisthepatientswhounderwentcoronarystenting≥3.5mmhadalowerrateofrestenosis(P<0.01).Collateralcirculationtotheobstructionsite,highmaximalinflationpressure,smokingandthelessminimallumendiameterafterPTCAmadetherateofrestenosishigherower(P<0.05).Multivariatelogisticregressionanalysisshowedthatcoronarystenting≥3.5mmhadalowrateofrestenosis,buthighmaximalinflationpressureandsmokingmadetherestenosisratehigher.ConclusionCoronarystentsize,maximalinflationpressureand.smokin
简介:AninformationtheorymethodisproposedtotesttheGrangercausalityandcontemporaneousconditionalindependenceinGrangercausalitygraphmodels.Inthegraphs,thevertexsetdenotesthecomponentseriesofthemultivariatetimeseries,andthedirectededgesdenotecausaldependence,whiletheundirectededgesreflecttheinstantaneousdependence.Thepresenceoftheedgesismeasuredbyastatisticsbasedonconditionalmutualinformationandtestedbyapermutationprocedure.Furthermore,fortheexistedrelations,astatisticsbasedonthedifferencebetweengeneralconditionalmutualinformationandlinearconditionalmutualinformationisproposedtotestthenonlinearity.Thesignificanceofthenonlinearteststatisticsisdeterminedbyabootstrapmethodbasedonsurrogatedata.Weinvestigatethefinitesamplebehavioroftheprocedurethroughsimulationtimeserieswithdifferentdependencestructures,includinglinearandnonlinearrelations.
简介:ThispapergivesanadditionaldefinitiontomultivariatetruncatedpowerT(x|x1,…,xn)whentheknotsx1,…,xndonotspanthewholespacetowhichx1,…,xnbelong.Bywhich,togetherwiththerecurrencerelation,thepracticalevaluationofT(x|x1,…,xn)willbeveryconvenientandefficient.Thesamediscussionisalsodonetoboxsplines.
简介:Inthispaper,rankfactorizationsandfactorleftprimefactorizationsarestudied.Theauthorsprovethatanypolynomialmatrixwithfullrowrankhasfactorleftprimefactorizations.Andforaclassofpolynomialmatrices,theauthorsgiveanalgorithmtodecidewhethertheyhaverankfactorizationsorfactorleftprimefactorizationsandcomputethesefactorizationsiftheyexist.
简介:THEGOODNESS-OF-FITTESTBYUSINGMULTIVARIATERANDOMWEIGHTING METHODTHEGOODNESS-OF-FITTESTBYUSINGMULTIVARIATERANDOMWEIGHTINGMETHOD...
简介:Letbethecollectionofm-timescontinuouslydifferentiableprobabilitydensitiesfonRdsuchthat丨Daf(x1)-Daf(x2)丨≤M‖x1-x2‖βforx1,x2∈Rd,[a]=m,whereDadenotesthedifferentialoperatordefinedbyDa=([a])/(x1a…xdad).UnderratherweakconditionsonK(x),thenecessaryandsufficientconditionsforsup丨n(x)-f(x)丨=0(((logn/n)λ/(d+3λ),λ=m+β,f∈arethat∫xaK(xi)dx=0for0<[a]≤m.Finallytheconvergencorateatapointisgiven.