Estimating the covariance matrix and the generalized variance under a symmetric loss

For estimating the power of a generalized variance under a multivariate normal distribution with unknown means, the inadmissibility of the best affine equivariant estimator relative to the symmetric loss is shown, and a class of improved estimators is given. The problem of estimating the covariance matrix is also discussed.     

Estimation of the Multivariate Normal Precision Matrix under the Entropy Loss

Let X ₁, , X _n (n > p) be a random sample from multivariate normal distribution N _p (, ), where R ^p and is a positive definite matrix, both and being unknown. We consider the problem of estimating the precision matrix ^–1. In this paper it is shown that for the entropy loss, the best lower-triangular affine equivariant minimax estimator of ^–1 is inadmissible and an improved estimator is explicitly constructed. Note that our improved estimator is obtained from the class of lower-triangular scale equivariant estimators.     

熵损失下协方差矩阵最佳仿射同变估计的改进(英)

设Ｘ１;…,Ｘｎ（ｎ＞ｐ）是来自多元正态分布Ｎｐ（μ,∑）的一个样本,其中μ∈Ｒ~ｐ,∑＞０均未知．本文在熵损失 Ｌ（ｓｕｍ　ｆｒｏｍ　ｔｏ　～,∑）＝ｔｒ（∑~－１,ｓｕｍ　ｆｒｏｍ　ｔｏ　～）－ｌｏｇ｜∑~－１ｓｕｍ　ｆｒｏｍ　ｔｏ～｜－ｐ下证明了协方差矩阵∑的最佳仿射同变估计是不容许的,且给出了其改进估计．     

对称熵损失下的一类指数族刻度参数估计

In this paper we investigate the estimator for the rth power of the scale parameter in a class of exponential family under symmetric entropy loss L（θ, δ） = v（θ/δ ＋ δ/θ - 2）. An exact form of the minimum risk equivariant estimator under symmetric entropy loss is given, and the minimaxity of the minimum risk equivariant estimator is proved. The results with regard to admissibility and inadmissibility of a class of linear estimators of the form cT（X） ＋ d are given, where T（X） Gamma（v, θ）.     

Improved estimation under Pitman's measure of closeness

Stein-type and Brown-type estimators are constructed for general families of distributions which improve in the sense of Pitman closeness on the closest (in a class) estimator of a parameter. The results concern mainly scale parameters but a brief discussion on improved estimation of location parameters is also included. The loss is a general continuous and strictly bowl shaped function, and the improved estimators presented do not depend on it, i.e., uniform domination is established with respect to the loss. The normal and inverse Gaussian distributions are used as illustrative examples. This work unifies and extends previous relevant results available in the literature.     

Estimation of the multivariate normal precision and covariance matrices in a star-shape model

In this paper, we introduce the star-shape models, where the precision matrix Ω (the inverse of the covariance matrix) is structured by the special conditional independence. We want to estimate the precision matrix under entropy loss and symmetric loss. We show that the maximal likelihood estimator (MLE) of the precision matrix is biased. Based on the MLE, an unbiased estimate is obtained. We consider a type of Cholesky decomposition of Ω, in the sense that Ω=Ψ′Ψ, where Ψ is a lower triangular matrix with positive diagonal elements. A special group , which is a subgroup of the group consisting all lower triangular matrices, is introduced. General forms of equivariant estimates of the covariance matrix and precision matrix are obtained. The invariant Haar measures on , the reference prior, and the Jeffreys prior of Ψ are also discussed. We also introduce a class of priors of Ψ, which includes all the priors described above. The posterior properties are discussed and the closed forms of Bayesian estimators are derived under either the entropy loss or the symmetric loss. We also show that the best equivariant estimators with respect to is the special case of Bayesian estimators. Consequently, the MLE of the precision matrix is inadmissible under either entropy or symmetric loss. The closed form of risks of equivariant estimators are obtained. Some numerical results are given for illustration. The project is supported by the National Science Foundation grants DMS-9972598, SES-0095919, and SES-0351523, and a grant from Federal Aid in Wildlife Restoration Project W-13-R through Missouri Department of Conservation.     

Admissibility and complete class results for the multinomial estimation problem with entropy and squared error loss

Let X ≡ (X₁, …, X_t) have a multinomial distribution based on N trials with unknown vector of cell probabilities p ≡ (p₁, …, p_t). This paper derives admissibility and complete class results for the problem of simultaneously estimating p under entropy loss (EL) and squared error loss (SEL). Let and f(x¦p) denote the (t − 1)-dimensional simplex, the support of X and the probability mass function of X, respectively. First it is shown that δ is Bayes w.r.t. EL for prior P if and only if δ is Bayes w.r.t. SEL for P. The admissible rules under EL are proved to be Bayes, a result known for the case of SEL. Let Q denote the class of subsets of of the form T = _j=1^kF_j where k ≥ 1 and each F_j is a facet of which satisfies: F a facet of such that F naF_jF ncT. The minimal complete class of rules w.r.t. EL when N ≥ t − 1 is characterized as the class of Bayes rules with respect to priors P which satisfy P( ⁰) = 1, ξ(x) ≡ ∫ f(x¦p) P(dp) > 0 for all x in {x : sup ^₀ f(x¦p) > 0} for some ⁰ in Q containing all the vertices of . As an application, the maximum likelihood estimator is proved to be admissible w.r.t. EL when the estimation problem has parameter space Θ = but it is shown to be inadmissible for the problem with parameter space Θ = ( minus its vertices). This is a severe form of “tyranny of boundary.” Finally it is shown that when N ≥ t − 1 any estimator δ which satisfies δ(x) > 0 x is admissible under EL if and only if it is admissible under SEL. Examples are given of nonpositive estimators which are admissible under SEL but not under EL and vice versa.     

Componentwise estimation of ordered parameters ofk (≥2) exponential populations

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.     

On the estimation of entropy

Motivated by recent work of Joe (1989,Ann. Inst. Statist. Math.,41, 683–697), we introduce estimators of entropy and describe their properties. We study the effects of tail behaviour, distribution smoothness and dimensionality on convergence properties. In particular, we argue that root-n consistency of entropy estimation requires appropriate assumptions about each of these three features. Our estimators are different from Joe's, and may be computed without numerical integration, but it can be shown that the same interaction of tail behaviour, smoothness and dimensionality also determines the convergence rate of Joe's estimator. We study both histogram and kernel estimators of entropy, and in each case suggest empirical methods for choosing the smoothing parameter.     

Existence of the uniformly minimum risk equivariant estimators of parameters in a class of normal linear models

In this paper, we study the existence of the uniformly minimum risk equivariant (UMRE) estimators of parameters in a class of normal linear models, which include the normal variance components model, the growth curve model, the extended growth curve model, and the seemingly unrelated regression equations model, and so on. The necessary and sufficient conditions are given for the existence of UMRE estimators of the estimable linear functions of regression coefficients, the covariance matrixV and (trV)^α, where α > 0 is known, in the models under an affine group of transformations for quadratic losses and matrix losses, respectively. Under the (extended) growth curve model and the seemingly unrelated regression equations model, the conclusions given in literature for estimating regression coefficients can be derived by applying the general results in this paper, and the sufficient conditions for non-existence of UMRE estimators ofV and tr(V) are expanded to be necessary and sufficient conditions. In addition, the necessary and sufficient conditions that there exist UMRE estimators of parameters in the variance components model are obtained for the first time.     

Improved variance estimation under sub-space restriction

For the linear regression model , we assume that for a given positive definite scale matrix , the error vector has a multivariate normal distribution and has the inverted Wishart distribution. For under an orthogonal sub-space restriction , we propose restricted unbiased, preliminary test and Stein-type estimators of variance of the error term, for when the scale of the inverse Wishart distribution is assumed to be unknown. We compare the weighted quadratic risks of the underlying estimators and propose dominance pictures for them.     

Non-parametric estimation of the generalized past entropy function with censored dependent data

The generalized past entropy function introduced by Gupta and Nanda (2002) is viewed as a dynamic measure of uncertainty in past life. This measure finds applications in modeling past life time data. In the present work we provide non-parametric kernel-type estimator for the generalized past entropy function based on censored data. Asymptotic properties of the estimator are established under suitable regularity conditions. Simulation studies are carried out using the Monte Carlo method.     

Estimation of the entropy of a multivariate normal distribution

Motivated by problems in molecular biosciences wherein the evaluation of entropy of a molecular system is important for understanding its thermodynamic properties, we consider the efficient estimation of entropy of a multivariate normal distribution having unknown mean vector and covariance matrix. Based on a random sample, we discuss the problem of estimating the entropy under the quadratic loss function. The best affine equivariant estimator is obtained and, interestingly, it also turns out to be an unbiased estimator and a generalized Bayes estimator. It is established that the best affine equivariant estimator is admissible in the class of estimators that depend on the determinant of the sample covariance matrix alone. The risk improvements of the best affine equivariant estimator over the maximum likelihood estimator (an estimator commonly used in molecular sciences) are obtained numerically and are found to be substantial in higher dimensions, which is commonly the case for atomic coordinates in macromolecules such as proteins. We further establish that even the best affine equivariant estimator is inadmissible and obtain Stein-type and Brewster–Zidek-type estimators dominating it. The Brewster–Zidek-type estimator is shown to be generalized Bayes.     

A sequence of improvements over the James-Stein estimator

In this article, we consider the problem of estimating a p-variate (p ≥ 3) normal mean vector in a decision-theoretic setup. Using a simple property of the noncentral chi-square distribution, we have produced a sequence of smooth estimators dominating the James-Stein estimator and each improved estimator is better than the previous one. It is also shown by using a technique of [5]. J. Multivariate Anal.36 121–126) that our smooth estimators can be dominated by non-smooth estimators.     

Minimax estimation of a multivariate normal mean under polynomial loss

Let X be an observation from a p-variate (p ≥ 3) normal random vector with unknown mean vector θ and known covariance matrix

. The problem of improving upon the usual estimator of θ, δ⁰(X) = X, is considered. An approach is developed which can lead to improved estimators, δ, for loss functions which are polynomials in the coordinates of (δ ? θ). As an example of this approach, the loss L(δ, θ) = |δ ? θ|⁴ is considered, and estimators are developed which are significantly better than δ⁰. When

is the identity matrix, these estimators are of the form $\text{δ(X) = (1 ? (}\text{b}\text{(d + |X|}{}^2\text{)}\text{))X}$.     

Minimax estimation of a multivariate normal mean under arbitrary quadratic loss

Let X be a p-variate (p ≥ 3) vector normally distributed with mean θ and known covariance matrix

. It is desired to estimate θ under the quadratic loss (δ ? θ)^tQ(δ ? θ), where Q is a known positive definite matrix. A broad class of minimax estimators for θ is developed.     

一种对称损失函数下一类指数分布族刻度参数的估计

在p,q对称熵损失函数L(θ,δ)=θp/δp+δq/θq-2(p,q0)下,研究了一类指数分布族c(x,n)θ-ve-T(x)/θ的刻度参数θ的Bayes估计与可容许估计,并应用积分变换定理证明了这两个估计具有不变性.     

Entropy loss and risk of improved estimators for the generalized variance and precision

Let the distributions of X(p×r) and S(p×p) be N(, I _r) and W _p(n, ) respectively and let them be independent. The risk of the improved estimator for || or {ei329-1} based on X and S under entropy loss (=d/|| –log(d/||)–1 or d||–log(d||)–1) is evaluated in terms of incomplete beta function of matrix argument and its derivative. Numerical comparison for the reduction of risk over the best affine equivariant estimator is given.Dedicated to Professor Yukihiro Kodama on his 60th birthday.     

A Note on the Best Invariant Estimator of a Distribution Function Under the Kolmogorov-Smirnov Loss

For the invariant decision problem of estimating a continuous distribution function with the Kolmogorov-Smirnov loss within the class of proper– distribution functions, it is proved that the sample distribution function is the best invariant estimator only for the sample size n = 1 and 2. Further it is shown that the best invariant estimator is minimax. Exact jumps of the best invariant estimator are derived for n 4.     

加权p,q对称熵损失下一类指数分布模型参数的估计

在由信息论中的熵演绎出的一种新损失一加权P,q对称熵损失L(θ,δ)=θ/Pδp+δq/qθq-2(ρ,q>0)下,研究了一类指数分布模型c(x,η)θ-νe-νe-T(x)/θ的参数θ的Bayes估计的一般形式与精确形式,讨论了参数θ的形如cT(X)+d的一类估计的可容许性与不可容许性,并应用积分变换定理证明了参数θ的Bayes估计与可容许估计具有不变性,