摘要:

Consider a mixed effects linear model Y = Xß+E, where Y € Rnlis a vector of observations, B € Rn1is mixed effects satisfying ß = A[d] + [d] with [d] N(0, o2[d]), [d] € Rk,1is fixed effects and [d] € Rp,lis random effects, which is uncorrelated withEN(0, o2V); X, A, [d] and V are known matrices, [d] and V are nonnegative definite. A random effects linear model and a fixed effects linear model are a special case of the above model. Necessary and sufficient conditions for a Linear estimator LY + b of S[d]+ Qß to be admissible in the class of linear estimators under the matrix risk are given

ISSN:0233-1888    

DOI:10.1080/02331888308802417

出版商:Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

An asymptotic upper bound is derived for the average expected squared error from polynomial regression. The bound is applied to determine guaranteed rates of convergence for estimation over certain function Classes, Two order selection techniques are shown to be optimal for selecting the number of terms to include in the estimator.

ISSN:0233-1888    

DOI:10.1080/02331888308802418

出版商:1993 Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

Summary. The general multivariate regression model with covariance components (the so called growth curve model) is studied in the paper. It is of the form Y = XBZ + [d], E[d] = 0, var(vec [d]) = W ® [d] =[d] We consider a parametric function [d]=[d]. There are derived conditions for unbiased and invariant estimation of parametric functions of the unknown parameters [d] and explicit formulas for locally best estimators under normality. The uniformly best estimator of the matrix W is also derived

ISSN:0233-1888    

DOI:10.1080/02331888308802419

出版商:Gordon and Breach Science Publishers S. A.    

年代:1993

数据来源: Taylor

摘要:

In this article, we consider the problem of estimation of variance covariance matrix a normally distributed population, when the population mean is known. We, specifically consider the maximum likelihood estimators derived under the known as well as unknown mean assumptions and test their relative merits under various loss functions. Definite results are obtained under entropy as well as squared error loss functions. Finally some open problems in this area are suggested.

ISSN:0233-1888    

DOI:10.1080/02331888308802420

出版商:Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

In this paper we present an exact Bayesian justification of the least-squares estimates for the functional relationships situation (Sprent, 1966) using the Reilly/Palatino-Leal Jeffreys' prior. It is shown that the profile posterior and the marginal posterior do not give conflicting information about the parameter of interest if an orthogonal parametrization (Cox and Reid, 1987) is adopted for the nuisance parameter.

ISSN:0233-1888    

DOI:10.1080/02331888308802421

出版商:Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

Accelerated life testing of an item under more severe than normal conditions is commonly used by industrial statisticians to reduce test time and costs. We consider, from the Bayesian point of view, the problem of accelerated iife tests when the parameter of interest is estimated via the orthogonal prameters (Cox and Reid, 1987) and a type II censoring mechanism for exponential survival data. One important advantage of this orthogonality is to obtain a closed form for the posterior distribution and its mode under the Jeffrey's prior for the parameters involved in the model. The method is illustrated by an example

ISSN:0233-1888    

DOI:10.1080/02331888308802422

出版商:Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

In this paper we investigate limiting properties of predictors of some finite population quantities. Both, the sample size and the population size are considered to become large. Limiting properties like consistency and asymptotic normality of the best linear unbiased predictors of the population total and of the finite population regression coefficient are investigated.

ISSN:0233-1888    

DOI:10.1080/02331888308802423

出版商:1993 Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

In this paper a new distance between probability measures is introduced. The definition of the distance is based on the use of ROC (Relative Operating Characteristic) curves to evaluate the accuracv of discriminant Statistics. Basic properties concerning this distance are studied, sucu as its relation with the error probabilities, with Fisher's separability measure and with the area under ROC curves.

ISSN:0233-1888    

DOI:10.1080/02331888308802424

出版商:Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

摘要:

Two classes of random distances (generated as contact distances or free path lengths) external to a stationary random closed set in Euclidean space are introduced. The censored distance distributions within bounded region are obtained. Unbiased estimators of the random distance distribution functions using only the information from inside the bounded region are constructed.

ISSN:0233-1888    

DOI:10.1080/02331888308802425

出版商:Gordon and Breach Science Publishers S.A.    

年代:1993

数据来源: Taylor

ISSN:0233-1888    

DOI:10.1080/02331888308802416

出版商:Gordon & Breach Sceince Publishers    

年代:1993

数据来源: Taylor