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.
Consider a standard row-column-exchangeable array X = (Xij : i,j ≥ 1), i.e., Xij = f(a, ξi, ηj, λij) is a function of i.i.d. random variables. It is shown that there is a canonical version of X, X′, such that X′, and α′, ξ1, ξ2,…, η1, η2,…, are conditionally independent given ∩n ≥ 1σ(Xij : max(i,j) ≥ n). This result is quite a bit simpler to prove than the analogous result for the original array X, which is due to Aldous.  相似文献   

2.
The celebrated U-conjecture states that under the Nn(0,In) distribution of the random vector X=(X1,…,Xn) in Rn, two polynomials P(X) and Q(X) are unlinkable if they are independent [see Kagan et al., Characterization Problems in Mathematical Statistics, Wiley, New York, 1973]. Some results have been established in this direction, although the original conjecture is yet to be proved in generality. Here, we demonstrate that the conjecture is true in an important special case of the above, where P and Q are convex nonnegative polynomials with P(0)=0.  相似文献   

3.
4.
In this paper we aim to construct adaptive confidence region for the direction of ξ in semiparametric models of the form Y=G(ξTX,ε) where G(⋅) is an unknown link function, ε is an independent error, and ξ is a pn×1 vector. To recover the direction of ξ, we first propose an inverse regression approach regardless of the link function G(⋅); to construct a data-driven confidence region for the direction of ξ, we implement the empirical likelihood method. Unlike many existing literature, we need not estimate the link function G(⋅) or its derivative. When pn remains fixed, the empirical likelihood ratio without bias correlation can be asymptotically standard chi-square. Moreover, the asymptotic normality of the empirical likelihood ratio holds true even when the dimension pn follows the rate of pn=o(n1/4) where n is the sample size. Simulation studies are carried out to assess the performance of our proposal, and a real data set is analyzed for further illustration.  相似文献   

5.
We consider the fixed design regression model Yi = g(ti) + ξi, i = 1, …, n, where ξi are (not necessarily i.i.d.) no variables, ti constitute the design points where nonrepeatable measurements are to be taken and Yi are the observations from which g and its derivatives are to be estimated. The dependency of the Integrated Mean Squared Error of two different types of kernel estimates on the design {t1, …, tn} is established. This allows the derivation of asymptotically optimal designs.  相似文献   

6.
Suppose that Y=(Yi) is a normal random vector with mean Xb and covariance σ2In, where b is a p-dimensional vector (bj),X=(Xij) is an n×p matrix. A-optimal designs X are chosen from the traditional set D of A-optimal designs for ρ=0 such that X is still A-optimal in D when the components Yi are dependent, i.e., for ii′, the covariance of Yi,Yi is ρ with ρ≠0. Such designs depend on the sign of ρ. The general results are applied to X=(Xij), where Xij∈{-1,1}; this corresponds to a factorial design with -1,1 representing low level or high level respectively, or corresponds to a weighing design with -1,1 representing an object j with weight bj being weighed on the left and right of a chemical balance respectively.  相似文献   

7.
We consider the problem of testing whether the common mean of a single n-vector of multivariate normal random variables with known variance and unknown common correlation ρ is zero. We derive the standardized likelihood ratio test for known ρ and explore different ways of proceeding with ρ unknown. We evaluate the performance of the standardized statistic where ρ is replaced with an estimate of ρ and determine the critical value cn that controls the type I error rate for the least favorable ρ in [0,1]. The constant cn increases with n and this procedure has pathological behavior if ρ depends on n and ρn converges to zero at a certain rate. As an alternate approach, we replace ρ with the upper limit of a (1−βn) confidence interval chosen so that cn=c for all n. We determine βn so that the type I error rate is exactly controlled for all ρ in [0,1]. We also investigate a simpler approach where we bound the type I error rate. The former method performs well for all n while the less powerful bound method may be a useful in some settings as a simple approach. The proposed tests can be used in different applications, including within-cluster resampling and combining exchangeable p-values.  相似文献   

8.
Komjáth in 1984 proved that, for each sequence (An) of analytic subsets of a Polish space X, if lim supnHAn is uncountable for every Hω[N] then ?nGAn is uncountable for some Gω[N]. This fact, by our definition, means that the σ-ideal [X]?ω has property (LK). We prove that every σ-ideal generated by X/E has property (LK), for an equivalence relation EX2 of type Fσ with uncountably many equivalence classes. We also show the parametric version of this result. Finally, the invariance of property (LK) with respect to various operations is studied.  相似文献   

9.
This paper proposes a constrained empirical likelihood confidence region for a parameter β0 in the linear errors-in-variables model: Yi=xiτβ0+εi,Xi=xi+ui,(1?i?n), which is constructed by combining the score function corresponding to the squared orthogonal distance with a constrained region of β0. It is shown that the coverage error of the confidence region is of order n−1, and Bartlett corrections can reduce the coverage errors to n−2. An empirical Bartlett correction is given for practical implementation. Simulations show that the proposed confidence region has satisfactory coverage not only for large samples, but also for small to medium samples.  相似文献   

10.
L. Foged proved that a weakly regular topology on a countable set is regular. In terms of convergence theory, this means that the topological reflection of a regular pretopology ξ on a countable set is regular. It is proved that this still holds if ξ is a regular σ-compact pretopology. On the other hand, it is proved that for each n<ω there is a (regular) pretopology ρ (on a set of cardinality c) such that k(RT)ρ>n(RT)ρ for each k<n and n(RT)ρ is a Hausdorff compact topology, where R is the reflector to regular pretopologies. It is also shown that there exists a regular pretopology of Hausdorff RT-order ?ω0. Moreover, all these pretopologies have the property that all the points except one are topological and regular.  相似文献   

11.
Let ?(η) be the class of positive random vectors T for which min1?i?nαiTi is IFRA (NBU) for all αi > 0, i = 1,…,n where n is an arbitrary positive integer. Characterizations of the classes ? and η are obtained and utilized to show that η is closed under convolution and that ? is closed under convolution provided one of the two convoluted vectors has independent components.  相似文献   

12.
We prove that every Borel bimeasurable mapping can be decomposed to a σ-discrete family of extended Borel isomorphisms and a mapping with a σ-discrete range. We get a new proof of a result containing the Purves and the Luzin-Novikov theorems as a by-product. Assuming an extra assumption on f, or that Fleissner's axiom (SCω2) holds, we characterize extended Borel bimeasurable mappings as those extended Borel measurable ones which may be decomposed to countably many extended Borel isomorphisms and a mapping with a σ-discrete range.  相似文献   

13.
A random balanced sample (RBS) is a multivariate distribution with n components Xk, each uniformly distributed on [-1,1], such that the sum of these components is precisely 0. The corresponding vectors lie in an (n-1)-dimensional polytope M(n). We present new methods for the construction of such RBS via densities over M(n) and these apply for arbitrary n. While simple densities had been known previously for small values of n (namely 2,3, and 4), for larger n the known distributions with large support were fractal distributions (with fractal dimension asymptotic to n as n→∞). Applications of RBS distributions include sampling with antithetic coupling to reduce variance, and the isolation of nonlinearities. We also show that the previously known densities (for n?4) are in fact the only solutions in a natural and very large class of potential RBS densities. This finding clarifies the need for new methods, such as those presented here.  相似文献   

14.
For simple point processes ξ on a Borel space S, we prove some approximations involving conditional distributions, given that ξ hits a small set B. Beginning with general versions of some classical limit theorems, going back to the pioneering work of Palm and Khinchin, we proceed to prove that, under suitable regularity conditions, the contributions to B and B c are asymptotically conditionally independent. We further derive approximations in total variation of reduced Palm distributions and show that, when ξ hits some small sets B 1,..., B n , the corresponding restrictions are asymptotically independent. Next we give general versions of the asymptotic relations P{ξ B > 0} ~ Eξ B and prove some ratio limit theorems for conditional expectations E[η | ξ B > 0], valid even when is not σ-finite and the Palm distributions may fail to exist.  相似文献   

15.
We find the asymptotic distribution of the OLS estimator of the parameters β and ρ in the mixed spatial model with exogenous regressors Yn=Xnβ+ρWnYn+Vn. The exogenous regressors may be bounded or growing, like polynomial trends. The assumption about the spatial matrix Wn is appropriate for the situation when each economic agent is influenced by many others. The error term is a short-memory linear process. The key finding is that in general the asymptotic distribution contains both linear and quadratic forms in standard normal variables and is not normal.  相似文献   

16.
We study the following model of hidden Markov chain: with (Xi) a real-valued positive recurrent and stationary Markov chain, and (?i)1?i?n+1 a noise independent of the sequence (Xi) having a known distribution. We present an adaptive estimator of the transition density based on the quotient of a deconvolution estimator of the density of Xi and an estimator of the density of (Xi,Xi+1). These estimators are obtained by contrast minimization and model selection. We evaluate the L2 risk and its rate of convergence for ordinary smooth and supersmooth noise with regard to ordinary smooth and supersmooth chains. Some examples are also detailed.  相似文献   

17.
Let Yn, n≥1, be a sequence of integrable random variables with EYn = xn1β1 + xn2β2 + … + xnpβp, where the xij's are known and βT = (β1, β2,…, βp) unknown. Let bn be the least-squares estimator of β based on Y1, Y2,…, Yn. Weak consistency of bn, n≥1, has been considered in the literature under the assumption that each Yn is square integrable. In this paper, we study weak consistency of bn, n≥1, and associated rates of convergence under the minimal assumption that each Yn is integrable.  相似文献   

18.
On the basis of a random sample of size n on an m-dimensional random vector X, this note proposes a class of estimators fn(p) of f(p), where f is a density of X w.r.t. a σ-finite measure dominated by the Lebesgue measure on Rm, p = (p1,…,pm), pj ≥ 0, fixed integers, and for x = (x1,…,xm) in Rm, f(p)(x) = ?p1+…+pm f(x)/(?p1x1 … ?pmxm). Asymptotic unbiasedness as well as both almost sure and mean square consistencies of fn(p) are examined. Further, a necessary and sufficient condition for uniform asymptotic unbisedness or for uniform mean square consistency of fn(p) is given. Finally, applications of estimators of this note to certain statistical problems are pointed out.  相似文献   

19.
I. Biswas 《Topology》2006,45(2):403-419
Let X be a nonsingular algebraic curve of genus g?3, and let Mξ denote the moduli space of stable vector bundles of rank n?2 and degree d with fixed determinant ξ over X such that n and d are coprime. We assume that if g=3 then n?4 and if g=4 then n?3, and suppose further that n0, d0 are integers such that n0?1 and nd0+n0d>nn0(2g-2). Let E be a semistable vector bundle over X of rank n0 and degree d0. The generalised Picard bundle Wξ(E) is by definition the vector bundle over Mξ defined by the direct image where Uξ is a universal vector bundle over X×Mξ. We obtain an inversion formula allowing us to recover E from Wξ(E) and show that the space of infinitesimal deformations of Wξ(E) is isomorphic to H1(X,End(E)). This construction gives a locally complete family of vector bundles over Mξ parametrised by the moduli space M(n0,d0) of stable bundles of rank n0 and degree d0 over X. If (n0,d0)=1 and Wξ(E) is stable for all EM(n0,d0), the construction determines an isomorphism from M(n0,d0) to a connected component M0 of a moduli space of stable sheaves over Mξ. This applies in particular when n0=1, in which case M0 is isomorphic to the Jacobian J of X as a polarised variety. The paper as a whole is a generalisation of results of Kempf and Mukai on Picard bundles over J, and is also related to a paper of Tyurin on the geometry of moduli of vector bundles.  相似文献   

20.
Let FX,Y(x,y) be a bivariate distribution function and Pn(x), Qm(y), n, m = 0, 1, 2,…, the orthonormal polynomials of the two marginal distributions FX(x) and FY(y), respectively. Some necessary conditions are derived for the co-efficients cn, n = 0, 1, 2,…, if the conditional expectation E[Pn(X) ∥ Y] = cnQn(Y) holds for n = 0, 1, 2,…. Several examples are given to show the application of these necessary conditions.  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