共查询到20条相似文献,搜索用时 31 毫秒
1.
The Fréchet distance between two multivariate normal distributions having means μX, μY and covariance matrices ΣX, ΣY is shown to be given by . The quantity d0 given by is a natural metric on the space of real covariance matrices of given order. 相似文献
2.
K.B. Athreya 《Statistics & probability letters》1983,1(3):147-150
Let X1, X2, X3, … be i.i.d. r.v. with E|X1| < ∞, E X1 = μ. Given a realization X = (X1,X2,…) and integers n and m, construct Yn,i, i = 1, 2, …, m as i.i.d. r.v. with conditional distribution for 1 ? j ? n. ( denotes conditional distribution given X). Conditions relating the growth rate of m with n and the moments of X1 are given to ensure the almost sure convergence of toμ. This equation is of some relevance in the theory of Bootstrap as developed by Efron (1979) and Bickel and Freedman (1981). 相似文献
3.
The following estimate of the pth derivative of a probability density function is examined: , where hk is the kth Hermite function and Σi = 1nhk(p)(Xi) is calculated from a sequence X1,…, Xn of independent random variables having the common unknown density. If the density has r derivatives the integrated square error converges to zero in the mean and almost completely as rapidly as O(n?α) and O(n?α log n), respectively, where . Rates for the uniform convergence both in the mean square and almost complete are also given. For any finite interval they are O(n?β) and , respectively, where . 相似文献
4.
Allan R Sampson 《Journal of multivariate analysis》1983,13(2):375-381
Let X1, …, Xp have p.d.f. g(x12 + … + xp2). It is shown that (a) X1, …, Xp are positively lower orthant dependent or positively upper orthant dependent if, and only if, X1,…, Xp are i.i.d. N(0, σ2); and (b) the p.d.f. of |X1|,…, |Xp| is TP2 in pairs if, and only if, In g(u) is convex. Let X1, X2 have p.d.f. . Necessary and sufficient conditions are given for f(x1, x2) to be TP2 for fixed correlation ?. It is shown that if f is TP2 for all ? >0. then (X1, X2)′ ~ N(0, Σ). Related positive dependence results and applications are also considered. 相似文献
5.
L.R. Haff 《Journal of multivariate analysis》1977,7(3):374-385
Let Sp×p ~ Wishart (Σ, k), Σ unknown, k > p + 1. Minimax estimators of Σ?1 are given for L1, an Empirical Bayes loss function; and L2, a standard loss function (Ri ≡ E(Li ∣ Σ), i = 1, 2). The estimators are , a, b ≥ 0, r(·) a functional on . Stein, Efron, and Morris studied the special cases and , for certain, a, b. From their work , a = k ? p ? 1, b = p2 + p ? 2; whereas, we prove . The reversal is surprising because a.e. (for a particular L2). Assume (compact) ? , the set of p × p p.s.d. matrices. A “divergence theorem” on functions Fp×p : → implies identities for Ri, i = 1, 2. Then, conditions are given for , i = 1, 2. Most of our results concern estimators with r(S) = t(U)/tr(S), U = p ∣S∣1/p/tr(S). 相似文献
6.
The probability measure of X = (x0,…, xr), where x0,…, xr are independent isotropic random points in n (1 ≤ r ≤ n ? 1) with absolutely continuous distributions is, for a certain class of distributions of X, expressed as a product measure involving as factors the joint probability measure of (ω, ?), the probability measure of p, and the probability measure of . Here ω is the r-subspace parallel to the r-flat η determined by X, ? is a unit vector in ω⊥ with ‘initial’ point at the origin [ω⊥ is the (n ? r)-subspace orthocomplementary to ω], p is the norm of the vector z from the origin to the orthogonal projection of the origin on η, and , where α is a scale factor determined by p. The probability measure for ω is the unique probability measure on the Grassmann manifold of r-subspaces in n invariant under the group of rotations in n, while the conditional probability measure of ? given ω is uniform on the boundary of the unit (n ? r)-ball in ω⊥ with centre at the origin. The decomposition allows the evaluation of the moments, for a suitable class of distributions of X, of the r-volume of the simplicial convex hull of {x0,…, xr} for 1 ≤ r ≤ n. 相似文献
7.
Morris L Eaton 《Journal of multivariate analysis》1976,6(3):422-425
Let Σ be an n × n positive definite matrix with eigenvalues λ1 ≥ λ2 ≥ … ≥ λn > 0 and let M = {x, y | x?Rn, y?Rn, x ≠ 0, y ≠ 0, x′y = 0}. Then for x, y in M, we have that and the inequality is sharp. If is a partitioning of Σ, let θ1 be the largest canonical correlation coefficient. The above result yields . 相似文献
8.
Kamal C Chanda 《Statistics & probability letters》1985,3(5):261-268
Let {Xt; t = 1, 2,…} be a linear process with a location parameter θ defined by Xt ? θ = Σ0∞grZt?r where {Zt; t = 0, ±1,…} is a sequence of independent and identically distributed random variables, with E∥Z1∥δ < ∞ for some δ > 0. If δ ? 1 we assume further than E(Z1) = 0. Let η = δ if 0 < δ < 2, and η = 2 if δ ? 2. Then assume that Σ0∞∥ gr ∥η < ∞. Consider the class of estimators given by is of the form cnt = Σp = 0sβnptp for some s ? 0. An attempt has been made to investigate the distributional properties of in large samples for various choices of βnp (0 ? p ? s), s, and the distribution of Z1 under the constraints Σ0∞rkgr = 0, 0 ? k ? q where q in an arbitrary integer, 0 ? q ? s. 相似文献
9.
C.S Withers 《Journal of multivariate analysis》1984,15(2):228-236
A bivariate Gaussian process with mean 0 and covariance is observed in some region Ω of R′, where {Σij(s,t)} are given functions and p an unknown parameter. A test of H0: p = 0, locally equivalent to the likelihood ratio test, is given for the case when Ω consists of p points. An unbiased estimate of p is given. The case where Ω has positive (but finite) Lebesgue measure is treated by spreading the p points evenly over Ω and letting p → ∞. Two distinct cases arise, depending on whether Δ2,p, the sum of squares of the canonical correlations associated with Σ(s, t, 1) on , remains bounded. In the case of primary interest as p → ∞, Δ2,p → ∞, in which case converges to p and the power of the one-sided and two-sided tests of H0 tends to 1. (For example, this case occurs when Σij(s, t) ≡ Σ11(s, t).) 相似文献
10.
For fixed p (0 ≤ p ≤ 1), let {L0, R0} = {0, 1} and X1 be a uniform random variable over {L0, R0}. With probability p let {L1, R1} = {L0, X1} or = {X1, R0} according as ; with probability 1 ? p let {L1, R1} = {X1, R0} or = {L0, X1} according as , and let X2 be a uniform random variable over {L1, R1}. For n ≥ 2, with probability p let {Ln, Rn} = {Ln ? 1, Xn} or = {Xn, Rn ? 1} according as , with probability 1 ? p let {Ln, Rn} = {Xn, Rn ? 1} or = {Ln ? 1, Xn} according as , and let Xn + 1 be a uniform random variable over {Ln, Rn}. By this iterated procedure, a random sequence {Xn}n ≥ 1 is constructed, and it is easy to see that Xn converges to a random variable Yp (say) almost surely as n → ∞. Then what is the distribution of Yp? It is shown that the Beta, (2, 2) distribution is the distribution of Y1; that is, the probability density function of Y1 is g(y) = 6y(1 ? y) I0,1(y). It is also shown that the distribution of Y0 is not a known distribution but has some interesting properties (convexity and differentiability). 相似文献
11.
Given a polynomial , we calculate a subspace Gp of the linear space 〈X〉 generated by the indeterminates which is minimal with respect to the property (the algebra generated by Gp, and prove its uniqueness. Furthermore, we use this result to characterize the pairs (P,Q) of polynomials P(X1,…,Xn) and Q(X1,…,Xn) for which there exists an isomorphism T:〈X〉 →〈X〉 that “separates P from Q,” i.e., such that for some k(1<k<n) we can write P and Q as and respectively, where . 相似文献
12.
A lower (upper) bound is given for the distribution of each dj, j = k + 1, …, p (j = 1, …, s), the jth latent root of AB?1, where A and B are independent noncentral and central Wishart matrices having with rank and Wp(n, Σ), respectively. Similar bound are also given for the distributions of noncentral means and canonical correlations. The results are applied to obtain lower bounds for the null distributions of some multivariate test statistics in Tintner's model, MANOVA and canonical analysis. 相似文献
13.
Hans G Weidner 《Journal of Number Theory》1976,8(1):99-108
Let g = (g1,…,gr) ≥ 0 and h = (h1,…,hr) ≥ 0, g?, h? ∈ , be two vectors of nonnegative integers and let λ ? , λ ≥ 0, λ ≡ 0 mod d, where d denotes g.c.d. (g1,…,gr). Define It is shown in this paper that Λ(λ) is periodic in λ with constant jump. If i? {1,…,r} is such that then holds true for all sufficiently large λ, λ ≡ 0 mod d. 相似文献
14.
A set {b1,b2,…,bi} ? {1,2,…,N} is said to be a difference intersector set if {a1,a2,…,as} ? {1,2,…,N}, j > ?N imply the solvability of the equation ax ? ay = b′; the notion of sum intersector set is defined similarly. The authors prove two general theorems saying that if a set {b1,b2,…,bi} is well distributed simultaneously among and within all residue classes of small moduli then it must be both difference and sum intersector set. They apply these theorems to investigate the solvability of the equations (, , , (where () denotes the Legendre symbol) and to show that “almost all” sets form both difference and sum intersector sets. 相似文献
15.
Noga Alon 《Journal of Combinatorial Theory, Series A》1985,40(1):82-89
Let X1, …, Xn be n disjoint sets. For 1 ? i ? n and 1 ? j ? h let Aij and Bij be subsets of Xi that satisfy |Aij| ? ri and |Bij| ? si for 1 ? i ? n, 1 ? j ? h, for 1 ? j ? h, for 1 ? j < l ? h. We prove that . This result is best possible and has some interesting consequences. Its proof uses multilinear techniques (exterior algebra). 相似文献
16.
Rasul A. Khan 《Stochastic Processes and their Applications》1979,9(2):207-215
Let be independent random variables, and set Wn = max(0,Wn-1 + Xn), W0 = 0, n ? 1. The so-called cusum (cumulative sum) procedure uses the first passage time T(h) = inf{n ? 1: Wn?h}for detecting changes in the mean μ of the process. It is shown that limh→∞ μET(h)/h = 1 if μ > 0. Also, a cusum procedure for detecting changes in the normal mean is derived when the variance is unknown. An asymptotic approximation to the average run length is given. 相似文献
17.
D.C Doss 《Journal of multivariate analysis》1979,9(3):460-464
The probability generating function (pgf) of an n-variate negative binomial distribution is defined to be [β(s1,…,sn)]?k where β is a polynomial of degree n being linear in each si and k > 0. This definition gives rise to two characterizations of negative binomial distributions. An n-variate linear exponential distribution with the probability function h(x1,…,xn) is negative binomial if and only if its univariate marginals are negative binomial. Let St, t = 1,…, m, be subsets of {s1,…, sn} with empty ∩t=1mSt. Then an n-variate pgf is of a negative binomial if and only if for all s in St being fixed the function is of the form of the pgf of a negative binomial in other s's and this is true for all t. 相似文献
18.
Let be a polynomial in the variables x1,…,xp with nonnegative real coefficients which sum to one, let A1,…,Ap be stochastic matrices, and let be the stochastic matrix which is obtained from ? by substituting the Kronecker product of An11,…,Anppfor each term Xn11·?·Xnpp. In this paper, we present necessary and sufficient conditions for the Cesàro limit of the sequence of the powers of to be equal to the Kronecker product of the Cesàro limits associated with each of A1,…,Ap. These conditions show that the equality of these two matrices depends only on the number of ergodic sets under and?or the cyclic structure of the ergodic sets under A1,…,Ap, respectively. As a special case of these results, we obtain necessary and sufficient conditions for the interchangeability of the Kronecker product and the Cesàro limit operator. 相似文献
19.
Allan R Sampson 《Statistics & probability letters》1984,2(2):77-81
A multivariate correlation ratio of a random vector upon a random vector is defined by where Λ, a fixed positive definite matrix, is related to the relative importance of predictability for the entries of . The properties of ηΛ are discussed, with particular attention paid to a ‘correlation-maximizing’ property. Given are applications of ηΛ to the elliptically symmetric family of distributions and the multinomial distribution. Also discussed is the problem of finding those r linear functions of that are most predictable (in a correlation ratio sense) from . 相似文献
20.
David S Jerison 《Journal of Functional Analysis》1981,43(1):97-142
For (x,y,t)∈n × n × , denote and . When α = n ? 2q, a represents the action of the Kohn Laplacian □b on q-forms on the Heisenberg group. For ?n < α < n, we construct a parametrix for the Dirichlet problem in smooth domains D near non-characteristic points of ?D. A point w of ?D is non-characteristic if one of X1,…, Xn, Y1,…, Yn is transverse to ?D at w. This yields sharp local estimates in the Dirichlet problem in the appropriate non-isotropic Lipschitz classes. The main new tool is a “convolution calculus” of pseudo-differential operators that can be applied to the relevant layer potentials, for which the usual asymptotic composition formula is false. Characteristic points are treated in Part II. 相似文献