首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 31 毫秒
1.
Zusammenfassung In den letzten Jahren erschien eine Reihe von Arbeiten, die sich systematisch mit Wahrscheinlichkeitsverteilungen auf topologischen Gruppen, Halbgruppen, topologischen RÄumen und topologischen linearen RÄumen beschÄftigten. Als besonders geeignet für eine topologische Wahrscheinlichkeitstheorie erwiesen sich hierbei die sogenannten straffen (tight) Wahrscheinlichkeitsverteilungen (vgl. Le Cam [3], Hildenbrand [11], Prochoeov [20], Varadarajan [25]).Die vorliegende Arbeit befa\t sich mit straffen Wahrscheinlichkeitsverteilungen im Raum D, dem topologischen Dualraum des Raumes D der auf der reellen Zahlengeraden definierten beliebig oft differenzierbaren Funktionen mit kompaktem TrÄger Tr .Der Ausgangspunkt für die Untersuchung von Zufallselementen mit Werten in linearen RÄumen, die nicht notwendig BanachrÄume sind, war wohl der von GELFAND [8] eingeführte Begriff des verallgemeinerten stochastischen Prozesses (VSP). Solange man bei einem solchen Proze\ Eigenschaften untersucht, die sich mit Hilfe seiner endlichdimensionalen Randverteilungen Q{1,...,n}, i D, beschreiben lassen, wird man sich wie im Fall eines gewöhnlichen stochastischen Prozesses natürlich die Frage stellen, ob ein geeigneter Standard-stichprobenraum existiert, etwa der Raum D, so da\ sich jeder VSP auffassen lÄ\t als Wahrscheinlichkeitsverteilung auf einem geeigneten hinreichend umfangreichen -Ring von Teilmengen des Raumes D. Die fundamentale Arbeit von MINLOS [18] gab hierzu die Lösung: Durch ein vertrÄgliches System endlichdimensionaler Wahrscheinlichkeitsverteilungen Q{1,...,n}, i D, mit gewissen Eigenschaften, die denen der Randverteilungen eines VSP entsprechen, lÄ\t sich auf dem SystemB der Zylindermengen des Raumes D eine sogenannte schwache Verteilung definieren, von der gezeigt wird, da\ sie -additiv ist. Durch EinschrÄnkung des Raumes der sogenannten Testfunktionen auf den metrisierbaren Teilraum D K{ D:Tr K, K kompakt in } von D lÄ\t sich dieses Ergebnis wie folgt verschÄrfen: Die durch ein vertrÄgliches System endlichdimensionaler Randverteilungen Q{1,...,n}, i D, mit entsprechenden Eigenschaften, auf dem System B K der Zylindermengen des Raumes DK definierte schwache Verteilung K ist straff bezüglich der schwachen Topologie (DK, DK) in DK.Die Frage nach der Gültigkeit einer entsprechenden VerschÄrfung für das Dualsystem >DD<, bzw. allgemeiner für ein Dualsystem E, F mit nicht notwendig metrisierbarem F, bildete den Gegenstand neuerer Untersuchungen, über deren Ergebnisse auf dem letzten Berkeley Symposium E. Mourier berichtete (vgl. [19]).Im ersten Kapitel der vorliegenden Arbeit des Verfassers wird demgegenüber eine Methode aufgezeigt, mit deren Hilfe, unter Verwendung des Minlosschen Satzes in seiner ursprünglichen Form, auf direktem Wege für das Dualsystem >D, D< der Nachweis gelingt, da\ eine schwache Verteilung auf B nicht nur -additiv, sondern automatisch straff ist (bzgl. der schwachen Topologie (D, D) in D) und sich somit eindeutig fortsetzen lÄ\t zu einer straffen Wahrscheinlichkeitsverteilung auf dem System 83 der Boreischen Mengen in D, welches den von den Zylindermengen erzeugten -Ring (B) umfa\t. Mit anderen Worten wird damit gezeigt, da\ man jeden VSP auffassen kann als straffe Wahrscheinlichkeitsverteilung auf den Boreischen Mengen in D. Wir sprechen dann auch von einer zufÄlligen Distribution.Im zweiten Kapitel betrachten wir spezielle zufÄllige Distributionen, nÄmlich Normal-verteilungen v, die aus Randverteilungen hervorgehen, welche n-dimensionale Normal-verteilungen sind, und beschÄftigen uns mit dem Problem der Äquivalenz und SingularitÄtzweier Normalverteilungen v1 und v2 in D. Für den Fall v1 = v, v2= vf 0, wo vf 0(Z) =v(Z – f0), ZB fD, zeigte DUDLEY [6], da\ entweder Äquivalenz oder SingularitÄt vorliegt, wobei er ein notwendiges und hinreichendes Kriterium für den Fall der Äquivalenz angibt. Aus der Theorie der gewöhnlichen stochastischen Prozesse ist nun bekannt, da\ die beiden Wahrschein-lichkeitsma\e, die zwei beliebigen Gau\schen Prozessen auf dem Raum ihrer Realisierungen entsprechen, entweder Äquivalent oder singular sind. Es lag deshalb nahe, nach einem Kriterium zu suchen, welches es einerseits gestattet, im Fall zweier beliebiger Normalverteilungen v1 und v2 in D zu entscheiden, wann Äquivalenz vorliegt, und welches andererseits die naheliegende Vermutung bestÄtigt, da\ für zwei Normalverteilungen in D dieselbe Alternative wie im eben zitierten klassischen Fall vorliegt. Dieses Problem wird gelöst, indem wir zeigen, da\ sich ein von Kallianfur-Oodaira [13] aufgestelltes Kriterium für die Äquivalenz zweier Normalverteilungen auf den Boreischen Mengen eines separablen Hilbertraumes auf den Distributionsraum D übertragen lÄ\t.Im dritten Kapitel beschÄftigen wir uns mit der Frage der Äquivalenz zweier beliebiger (nicht notwendig normaler) Wahrscheinlichkeitsverteilungen in D.Abschlie\end möchte der Autor Herrn Professor Dr. K. Krickeberg (Heidelberg) für die Anregung zu dieser Arbeit sowie für die Unterstützung wÄhrend ihrer Durchführung herzlich danken.  相似文献   

2.
Summary In this paper we study the asymptotic behaviour (as h) of the solutions of minimum problems for the functional [¦Du¦2+g(x, u)]dx with bilateral obstacles of the type huh, where h and h are sequences of arbitrary functions fromR n into ¯R.  相似文献   

3.
Summary Given a Markov chain (X n ) n0, random times are studied which are birth times or death times in the sense that the post- and pre- processes are independent given the present (X –1, X ) at time and the conditional post- process (birth times) or the conditional pre- process (death times) is again Markovian. The main result for birth times characterizes all time substitutions through homogeneous random sets with the property that all points in the set are birth times. The main result for death times is the dual of this and appears as the birth time theorem with the direction of time reversed.Part of this work was done while the author was visiting the Department of Mathematics, University of California at San DiegoThe support of The Danish Natural Science Research Council is gratefully acknowledged  相似文献   

4.
5.
Let 1, 2, ... be a sequence of i.i.d. random variables with positive mean and finite variance and letr(b), b0, be real numbers tending to 0 asb . Definings n=1+...+n andS n=Sn(b)=sn+r(b)n, the stopping time =(b)=inf {n>/1:Sn >b} whereb=b(b) , will be considered with special regard to the excess over the boundaryR b=s+r(b)–b. It turns out that the limiting distribution ofR b is the same as in the caser(b)0 for allb. Proving this, Blackwell's renewal theorem and its integral version have to be established first in the above stated situation. Finally, an expansion ofE to vanishing terms asb will be provided and applied to some examples arising in economics.
Zusammenfassung Seien 1, 2, ... unabhängige identisch verteilte Zufallsgrößen mit positivem Erwartungswert und endlicher Varianz sowier(b), b0, reelle Zahlen mitr(b)0 für b. Sei ferners 1, s2, ... der zugehörige Summenprozeß,S n= Sn(b)=sn+r(b)n fürn1 und =(b)=inf {n1: Sn>b, wobeib=b(b) fürb . Es wird gezeigt, daß die asymptotische Verteilung des ExzessesR b=s +r(b)b mit der im Fallr(·)0 übereinstimmt. Dazu werden sowohl das Blackwellsche Erneuerungstheorem als auch seine Integralversion in der vorher beschriebenen parameterabhängigen Situation geeignet formuliert und bewiesen. Als Folgerung ergibt sich dann eine asymptotische Entwicklung vonE(b) fürb bis zu Termen o(1). Anh- and einiger Beispiele aus dem ökonomischen Bereich wird schließlich noch aufgezeigt, wo Approximationen fürE(b) von Interesse sein können.
  相似文献   

6.
A study is made of the degeneracy of multidimensional dispersion laws (k) that increase unboundedly as k and satisfy some additional conditions. Under the assumption that the corresponding degeneracy functionf(k) satisfies a certain condition [Eq. (4)], it is shown that only two-dimensional dispersion laws of the form (p,q)=p 3(q/p)+cp(q/p)(|p|,|q|1), wherep(q/p)=f(p, q) is the corresponding unique degeneracy function, can be degenerate with respect to a 12 process. Some conditions that the function () must satisfy are obtained. The explicit form of a degenerate dispersion law with functionp 3(q/p) of polynomial form is found.Institute of Physics, Georgian Academy of Sciences. Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 95, No. 1, pp. 20–33, April, 1993.  相似文献   

7.
This paper deals with adapting Runge-Kutta methods to differential equations with a lagging argument. A new interpolation procedure is introduced which leads to numerical processes that satisfy an important asymptotic stability condition related to the class of testproblemsU(t)=U(t)+U(t–) with , C, Re()<–||, and >0. Ifc i denotes theith abscissa of a given Runge-Kutta method, then in thenth stept n–1t n :=t n–1+h of the numerical process our interpolation procedure computes an approximation toU(t n–1+c i h–) from approximations that have already been generated by the process at pointst j–1+c i h(j=1,2,3,...). For two of these new processes and a standard process we shall consider the convergence behaviour in an actual application to a given, stiff problem.  相似文献   

8.
We obtain a complete solution of the problem of the maximum of the fourth diameter in the family of continua with capacity 1. Let E(o, ei, e–i). 0<i, e–i; H(=cap E(o, ei, e–i). Let C() be the common point of three analytic arcs which form E(o, ei, e–i). One shows that the indicated maximum is realized by the continuum ={z:H(0)z 2E(o, ei, e–i)} where 0, o<0z ei z+C ( is a real and C is a complex constant). One finds the value of the required maximum. The paper contains a brief exposition of the proof of this result.Translated from Zapiski Nauchnykh Seminarov Leningradskogo Otdeleniya Matematicheskogo Instituta im. V. A. Steklova AN SSSR, Vol. 59, pp. 60–79, 1976.  相似文献   

9.
Consider the Product Rate Variation problem. Given n products 1,...,i,...,n, and n positive integer demands d 1,..., di,...,dn. Find a sequence =1,...,T, T = i=1 n d i, of the products, where product i occurs exactly d i times that always keeps the actual production level, equal the number of product i occurrences in the prefix 1,..., t, t=1,...,T, and the desired production level, equal r i t, where r i=di/T, of each product i as close to each other as possible. The problem is one of the most fundamental problems in sequencing flexible just-in-time production systems. We show that if is an optimal sequence for d 1,...,di,...,dn, then concatenation m of m copies of is an optimal sequence for md 1,..., mdi,...,mdn.  相似文献   

10.
A topological spaceX whose topology is the order topology of some linear ordering onX, is called aninterval space. A space in which every closed subspace is homeomorphic to a clopen subspace, is called aCO space and a space isscattered if every non-empty subspace has an isolated point. We regard linear orderings as topological spaces, by equipping them with their order topology. IfL andK are linear orderings, thenL *, L+K, L · K denote respectively the reverse ordering ofL, the ordered sum ofL andK and the lexicographic order onL x K (so · 2=+). Ordinals are considered as linear orderings, and cardinals are initial ordinals. For cardinals , l 0, letL(K,)=K+1+*.Theorem: Let X be a compact interval scattered space. Then X is a CO space if and only if X is homeomorphic to a space of the form +1+1 L(K i i), where is any ordinal, n , for every ii,i are regular cardinals and Kii, and if n>0, then max({Ki:i相似文献   

11.
Summary Let (xini, y i be a sequence of independent identically distributed random variables, where x i R p and y i R, and let R p be an unknown vector such that y i =x i +u i (*), where u i is independent of x i and has distribution function F(u/), where >0 is an unknown parameter. This paper deals with a general class of M-estimates of regression and scale, ( *,*), defined as solutions of the system: , where r= (y i x i 1*/)*, with R p ×RR and RR. This class contains estimators of (, ) proposed by Huber, Mallows and Krasker and Welsch. The consistency and asymptotic normality of the general M-estimators are proved assuming general regularity conditions on and and assuming the joint distribution of (x i , y i ) to fulfill the model (*) only approximately.  相似文献   

12.
Summary We discuss in this paper a non-homogeneous Poisson process A driven by an almost periodic intensity function. We give the stationary version A * and the Palm version A 0 corresponding to A *. Let (T i ,i) be the inter-point distance sequence in A and (T i 0 ,i) in A 0. We prove that forj, the sequence (T i+j,i) converges in distribution to (T i 0 ,i). If the intensity function is periodic then the convergence is in variation.  相似文献   

13.
We consider depth first search (DFS for short) trees in a class of random digraphs: am-out model. Let i be thei th vertex encountered by DFS andL(i, m, n) be the height of i in the corresponding DFS tree. We show that ifi/n asn, then there exists a constanta(,m), to be defined later, such thatL(i, m, n)/n converges in probability toa(,m) asn. We also obtain results concerning the number of vertices and the number of leaves in a DFS tree.  相似文献   

14.
Conditions are found which must be imposed on a function g(x) in order that M g(1+2+ + v < if M g(i) < and M g(v) < ,, 1, 2, , n, ... being non-negative and independent, being integral, and {i} being identically distributed. The result is applied to the theory of branching processes.Translated from Matematicheskie Zametki, Vol. 3, No. 4, pp. 387–394, April, 1968.  相似文献   

15.
A topological space X whose topology is the order topology of some linear ordering on X, is called an interval space. A space in which every closed subspace is homeomorphic to a clopen subspace, is called a CO space. We regard linear orderings as topological spaces, by equipping them with their order topology. If L and K are linear orderings, then L *, L+K, L·K denote respectively the reverse orderings of L, the ordered sum of L and K and the lexicographic order on L×K (so ·2=+ and 2·=). Ordinals are considered as linear orderings, and cardinals are initial ordinals. For cardinals , 0, let L(, )= + 1 + * . Main theorem. Let X be a compact interval space. Then X is a CO space if and only if X is homeomorphic to a space of the form + 1 + i L( i , i ), where is any ordinal, n, for every ii, i are regular cardinals and i i, and if n>0, then max({ i: i}) · . This first part is devoted to show the following result. Theorem: If X is a compact interval CO space, then X is a scattered space (that means that every subspace of X has an isolated point).Supported by the Université Claude-Bernard (Lyon-1), the Ben Gurion University of the Negev, and the C.N.R.S.: UPR 9016Supported by the City of Lyon  相似文献   

16.
For a fixed unit vectora=(a 1,...,a n )S n-1, consider the 2 n sign vectors=(1,..., n ){±1{ n and the corresponding scalar products·a = n i=1 = i a i . The question that we address is: for how many of the sign vectors must.a lie between–1 and 1. Besides the straightforward interpretation in terms of the sums ±a 2 , this question has appealing reformulations using the language of probability theory or of geometry.The natural conjectures are that at least 1/2 the sign vectors yield |.a|1 and at least 3/8 of the sign vectors yield |.a|<1 (the latter excluding the case when |a i |=1 for somei). These conjectured lower bounds are easily seen to be the best possible. Here we prove a lower bound of 3/8 for both versions of the problem, thus completely solving the version with strict inequality. The main part of the proof is cast in a more general probabilistic framework: it establishes a sharp lower bound of 3/8 for the probability that |X+Y|<1, whereX andY are independent random variables, each having a symmetric distribution with variance 1/2.We also consider an asymptotic version of the question, wheren along a sequence of instances of the problem satisfying ||a||0. Our result, best expressed in probabilistic terms, is that the distribution of .a converges to the standard normal distribution, and in particular the fraction of sign vectors yielding .a between –1 and 1 tends to 68%.This research was supported in part by the Institute for Mathematics and its Applications with funds provided by the National Science Foundation.  相似文献   

17.
We consider the estimation of ordered parameters ofk ( 2) exponential distributions by improving upon the usual estimators. TheBrewsterzidek technique is used to find sufficient conditions for an estimator of i and/or i (i=1,...,k), to be inadmissible with respect to the MSE criterion where i and i are the location and scale parameters respectively of thei-th exponential population. Using these sufficient conditions improved estimators of i and/or i (i=1,...,k) are obtained.  相似文献   

18.
Let denote a bipartite distance-regular graph with diameter D 4, valency k 3, and distinct eigenvalues 0 > 1 > ··· > D. Let M denote the Bose-Mesner algebra of . For 0 i D, let E i denote the primitive idempotent of M associated with i . We refer to E 0 and E D as the trivial idempotents of M. Let E, F denote primitive idempotents of M. We say the pair E, F is taut whenever (i) E, F are nontrivial, and (ii) the entry-wise product E F is a linear combination of two distinct primitive idempotents of M. We show the pair E, F is taut if and only if there exist real scalars , such that i + 1 i + 1 i – 1 i – 1 = i ( i + 1 i – 1) + i ( i + 1 i – 1) + (1 i D – 1)where 0, 1, ..., D and 0, 1, ..., D denote the cosine sequences of E, F, respectively. We define to be taut whenever has at least one taut pair of primitive idempotents but is not 2-homogeneous in the sense of Nomura and Curtin. Assume is taut and D is odd, and assume the pair E, F is taut. We show
for 1 i D – 1, where = 1, = 1. Using these equations, we recursively obtain 0, 1, ..., D and 0, 1, ..., D in terms of the four real scalars , , , . From this we obtain all intersection numbers of in terms of , , , . We showed in an earlier paper that the pair E 1, E d is taut, where d = (D – 1)/2. Applying our results to this pair, we obtain the intersection numbers of in terms of k, , 1, d, where denotes the intersection number c 2. We show that if is taut and D is odd, then is an antipodal 2-cover.  相似文献   

19.
Tomasz Łuczak 《Order》1991,8(3):291-297
Let =(n,p) be a binary relation on the set [n]={1, 2, ..., n} such that (i,i) for every i and (i,j) with probability p, independently for each pair i,j [n], where i<j. Define as the transitive closure of and denote poset ([n], ) by R(n, p). We show that for any constant p probability of each first order property of R(n, p) converges as n .  相似文献   

20.
Nous donnons une caractérisation des domaines DX pour lesquels la fonction extrémale relative *(,E,D) a la propriété de stabilité pour tout ED, i.e. lim k*(,E,D k )=*(,E,D), ED. Ensuite, nous étudions la relation entre cette propriété et les enveloppes pluripolaires. Nous concluons par quelques remarques sur la propriété de stabilité lim k*(,E k ,D)=*(,E,D).  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号