STRONGLY CONSISTENT ESTIMATION FOR A MULTIVARIATE LINEAR RELATIONSHIP MODEL WITH ESTIMATED COVARIANCES MATRIX

1 IntroductionSuppose that the true values Of a set of variables satisfy exact linear relationships, but theyare unobserved. We can Obtain the Observed values which are the sums of the true values anderrors of measurement. The model is described as follows. A set of rmdimensional vectors {xh}satisfyB'xk or = 0, k = 1,2,'' t (1.1)where B is an m x p parameter mains with rank p < m and o is a p x 1 parameter vector. Weobserved {Zk}, which are m x 1 random vectors and satisfyZk = ac ed, …     

Characteristic Properties of Perspective Sets for Finite Stochastic Decision

1.INTRODUCTIONConsequenceswilloccuruncertainlyilldecisionwithprobability,becallseoftheuncertainticofstates.Naturallyweshouldcollsidertheuncertaillt}'whelltheutilityofcollsequen"esis'liscussed.1\reindicatetilesetofally{CI.CZ.''.C,,}withtilecorrespondillgsetI)ofprobabilities{pl,p2,''.p,,}sucllthat'Epi=1.SettingP=(pl.CI;I)2.C2;'':p,,.C,,),]=[wecallPasaperspective.ThesetofallPerspectivederivedby'{CI.C2,'',C,,}isilldicatedby9,thatis,RatiollalBehaviorAxiomsassureth…     

L_1 DECONVOLUTION FOR A STATIONARY NON-MINIMUM PHASE AR(p) MODEL

This paper provides the theoretical fundamentals of L_1 deconvolution for thestationary time series{y_n}which satisfies a non-minimum phase AR(p)model.An algorithmof the L_1 deconvolution and its convergency are analyzed in detail.Both theoretical analysisand numerical examples show that this kind of deconvolution is especially appropriate forthe case of P(x_n=0)≠0,where{x_n}is the system input resulting in{y_n};and such a caseis just interesting in seismic signal processing.     

THE ESTIMATE AND TEST IN THE SYMMETRICAUXILIARY INFORMATION

1.IntroductionLetX,XI,X2,''lXui.i.d.~FO.TheempiricaldistributionFn=Z::,6x./niswellknowntobethenonparametricmaximumlikelihoodestimateofFObasedonXI,X2,'')Xu.Here6:denotesapointmassatx.ThelikelihoodfunctionthatFimaximizesisL(G)=fi:=,G{xi},whereG{xi}istheprobabilityoftheset{xi}underG,xiistheobservedvalueofXiandGisanyprobabilitymeasureonCP.WedefinetheempiricallikelihoodfunctionnL(G)=fiG{xi},whereG<相似文献   

MAXIMUM PRINCIPLE FOR OPTIMAL CONTROLPROBLEM OF FULLY COUPLEDFORWARD-BACKWARD STOCHASTIC SYSTEMS

1.IntroductionLet(fi,F,P)beaprobabilityspaceand{Bt}tZobead-dimensionalBrownianmotioninthisspace.Let{R}tZobethenaturalfiltrationofthisBrownianmotion.Weconsiderthefollowingfullycoupledforward-backwardstochasticsystems:where(x,y,z)takesvaluesinR"xacxRTnxd.LetUbeanonemptyconvexsubsetofR',Anelementofadiscalledanadmissiblecontrol.Wecandefinethefollowingcostfunction:TheoptimalcontrolproblemistominimizethecostfunctionJ(v(.))overadmissiblecontrols.Anadmissiblecontrolu(.)iscalledanoptimalcontrol…     

数字信号处理和时序分析(续)

2.4 数字系统的分类 一个系统(或滤波器或运算子)不过是被称做系统输入{x_n}和系统输出{y_n}的两个量之间的特定关系: {x_n}→{y_n} 其中符号“→”表示输入{x_n}引起输出{y_n)。如果输入和输出都是序列,该系统就叫做数字系统或数字滤波器。 当且仅当系统能满足下列特性时,系统     

基于伪布尔函数的n人对策模糊延拓方法

基于伪布尔函数,给出了不同于传统多重线性延拓和cornet延拓的一种n人对策的模糊延拓方法.由于n人对策中任意的联盟可由它的特征向量来等价的表示,利用伪布尔函数, 将n人对策从集合{0, 1}^{n}延拓到[0,1]^{n}上, 详细讨论了这种模糊延拓的性质,研究它与经典n人对策的Shapley值之间的关系.     

TREE DECOMPOSITIONS OF MULTIGRAPHS

1.IntroductionInthispaperweconsideronlyfiniteundirectedlooplessgraphs.ForasubgraphHofG,E(H)andP(H)denotethenumberofedgesandthenUmberofcyclesinHrespectively.GiventwodisjointsubsetsXandYofV(G),wewriteEG(X,Y)={acEE(G)IxEX,y6Y}.IfTisatreeinGande=acEG--E(T)with{u,v}CV(T),thenT econtainsauniquecycle.WedenotethiscyclebyC(T,e).Aforestiscalledak-tree-forestifitconsistsofexactlykpairwisedisjointtrees.ForagraphG,ifE(G)canbepartitionedilltoseveralpairwisedisjointsetsas{EI,EZ,...,EI}such…     

Estimation of Parameter in the Case of Zero-Failure Data

l IntroductionFrom the article [l] presenting, the zero-failure data research has been twenty years longand more attention is paid by researchers all over the world. We have got someachievements already. The devclopment of these researches can be found in article L2j'In article [3j, a method dealing with zero-failure data, that is, distribution curvesmethod, is given and has more values. The basic thought is: at first estimation the failureprobability pi = p {Twtt. } at moment l,, then appro…     

NOTICE ON ADDED-FACTOR OF MULTIVARIATE GOULD-HSU INVERSION

1. IntroductionIn 1973, W. H. Gould and L. C. Hsu [1] obtained a pair of reciprocal relations as follows.Theorem 1 Let {al}gr and {hi}gr be two sequences of complex number. Define thatand let that of (x, 0) = 1. Then there exists a pair of reciprocal relations as followsandwhere (:) denotes the oofinary binomial coefficients.This result is often called Gould-Hsu inversion formula in the literature, which turns outto be of great importance in the transformation of illtegral sequencest the c…     

带有B&L转换规则的Vpverline{X}控制图的经济设计

在已有Vpverline{X}控制图基础上, 结合B&L转换规则,构建了新的 控制图经济模型, 给出了相应的费用函数. 将其与现有Vpverline{X}控制图进行了比较分析, 发现在缩短报警时间、节约费用方面, 所构建的模型有着明显的优势. 此外,运用实际案例验证了模型的节约效果,并运用遗传算法、析因试验设计和多元回归方法,对过程参数和费用参数进行了敏感性分析.     

ENDOMORPHISMS OF LIE ALGEBRA F[t]d／dt

Let F be a field of characteristic zero. W_n =is the Witt algebra over F, W_n~ =is Lie subalgebra of W_n. It is well known both W_n and W_n~ are simple infinite dimensional Lie algebra. In Zhao's paper, it was conjectured thatEnd(W_n~ ) - {0} = Aut(W_n~ ) and it was proved that the validity of this conjecture im-plies the validity of the well-known Jacobian conjecture. In this short note, we check theconjecture above for n = 1. We show End(W_1~ ) - {0} = Aut(W_1~ ).     

THE NEIGHBORHOOD INTERSECTIONS OF ESSENTIAL SETS AND HAMILTONICITY OFGRAPHS

ThisresearchissupportedbytheNationalNaturalScienceFOundationofChina.1.IntroductionInthispaper,theterminologyandnotationnotdefinedwillfollow[1],andweconsideronlysimplefinitegraphs.Le Gbeagraph.AnindependentsetYinGiscalledanessentialindependentset(oressentialsetforsimplicity)ifthereis{UI,y2}CYsuchthatdist(yl,y2)~2.Lett>1beaninteger.DenoteIt(G)={YIYisanindependetsetofG,IYI~t}.It*(G)={YIYisanessentialsetofG,IYI~t}.LetYgV(G),andIYI~t.FOreachiE{0,l,2,'',t},denoteSt(Y)={yivEV(G),…     

混合基表示下的大数比较算法及其最优性

本文给出了大数在混合基表示下的比较算法,主要结论为:设X=b_(n 1)P_nP_(n-1)…P_1 … b_2p_1 b_1,其中{p_i}两两互素,则对于计算{b_i}的任何算法,计算b_(n 1)所花的时间大于计算b_1,b_2,… ,b_n所花时间的总和.从而利用计算b_1,b_2,…,b_n的最优算法进行两个大数的比较,本文提出的比较算法是最优的.     

THE CAUCHY PROBLEM FOR GENERAL COUPLEDMAXWELL-SCHR■DINGER EQUATIONS (II)

1.IntroductionandMainResultIn{11westudiedthelocalwell-posednessfortheCauchyproblemforgeneralcoupledMaxwelLSchr6dingerequations.Thisisacontinuationof[1].ItisdevotedtothestudyoftheglobalexistenceofCauchyproblemforgeneralcoupledMaxwell--SchrodlugerequaticusunderLorentzgaugewhenthespatialdimensionsareoneortwobyapplyingthequasi-energymethodwhichwasintroducedbyV.Moncerieft2].Basingon[1],theCauchyproblemforgeneralcoupledMaxweif--Schr6dingerequaticusnuderLorentzgaugecanbewrittenaswherexER",p,u…     

Robust极点配置及其优化算法

本文提出了一个线性系统极点配置的有效算法,首先利用正交变换将能控系统简化成上Hessenberg能控标准型,然后利用(A,B)特征子空间找出闭环系统关于指定复数集所有可能的特征向量,从中选取一组线性无关的向量{x_1,…x_n}作为闭环系统的特征向量,使得max{c_k,k∈n}极小,最后由x={x_1,…x_n}找出反馈阵F,使得闭环系统(A+BF,B)的极点集等于L。     

ON THE CONDITION FOR CONSISTENCY OF RESAMPLING ESTIMATORS OF VARIANCE OF TRIMMED L-STATISTICS

1.IntroductionLetXI,X2,''tX.beani.i.d.randomsamplewithcommondistributionfunction(c.d.f.)F(unknown),andFibetheempiricaldistributionfunction(e.d.f.)basedonthesample.ConsiderthefollowingfunctionalofFwhereF--'(t)ginf{x:F(x)2t}istheinversefunctionofF(inversef…     

ON THE ONE-SIDED LOGARITHMIC LAW FOR ARRAYS

1.IntroductionLet{X.,n2l}beasequenceofindependentandidenticallydistributed(i.i.d.)randomvariables.TheHartmanandWintner,sllllawoftheiteratedlogarithmstatesthatifEXI~0andEX7=1Moreover,itisprovedin[2]and[3]that(1)holdsifandonlyifEXI~0andEX7=1.Nowlet{Xnk,k=1,2,''tntn~1,2,'.}beanarrayofi.i.d.randomvariables.FOreveryn21setS.~Z:--,X.k'ThealmostsureconvergenceofSahasbeenstlldiedbyHu,MoriczandTaylor[41,HuandWeb.rls]andQi[6].HuandWeber[5]showedundertheconditionsEXll=0,EX7,=1andEXt,相似文献   

A ROBUST SQP METHOD BASED ON A SMOOTHING APPROXIMATE PENALTY FUNCTION FOR INEQUALITY CONSTRAINED OPTIMIZATION

1 IntroductionConsider the following inequality constrained optiInization problem:Inin f(x)xeRn(1)s.t. g(x) 5 0where f: R1'-R, g: R"-Rm are cofltinuousIy differentiable functions.SQP method for soIving problem(1) is to gellerate a sequence {xk} converging…     

PRE-IMAGE ENTROPY OF NONAUTONOMOUS DYNAMICAL SYSTEMS*

The authors define and study topological pre-image entropy for the non-autonomous discrete dynamical systems given by a sequence {fi}i=1^∞ of continuous self-maps of a compact topological space. The basic properties and the invariant with respect to equiconjugacy of pre-image entropy for the non-autonomous discrete dynamical systems are obtained.