For a seemingly Unrelated regression system with the assumption of normality,a necessary and sufficient condition for the existence of the Uniformly Minimum Risk Unbiased (UMRU)estimator of regression coefficients und...For a seemingly Unrelated regression system with the assumption of normality,a necessary and sufficient condition for the existence of the Uniformly Minimum Risk Unbiased (UMRU)estimator of regression coefficients under strictly convex loss is obtained;it is proved that any unbiased estimator can not improve the least squares estimator;it is also shown that no UMRU estimator exists under missing observations.展开更多
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 ...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.展开更多
Throughout this note, the following notations are used. For matrices A and B,A】B means that A-B is positive definite symmetric, A×B denotes the Kroneckerproduct of A and B R（A）, A’ and A<sup>-</sup&g...Throughout this note, the following notations are used. For matrices A and B,A】B means that A-B is positive definite symmetric, A×B denotes the Kroneckerproduct of A and B R（A）, A’ and A<sup>-</sup> stand for the column space, the transpose andany g-inverse of A, respectively; P<sub>A</sub>=A（A’A）<sup>-</sup>A’;for s×t matrix B=（b<sub>1</sub>…b<sub>t</sub>）,vec（B） de-notes the st-dimensional vector （b<sub>1</sub>′b<sub>2</sub>′…b<sub>t</sub>′）′, trA stands for the trace of the square ma-trix A.展开更多
Opting to follow the computing-design philosophy that the best way to reduce power consumption and increase energy efficiency is to reduce waste, we propose an architecture with a very simple ready-implementation by u...Opting to follow the computing-design philosophy that the best way to reduce power consumption and increase energy efficiency is to reduce waste, we propose an architecture with a very simple ready-implementation by using an NComputing device that can allow multi-users but only one computer is needed. This intuitively can save energy, space as well as cost. In this paper, we propose a simple and realistic NComputing architecture to study the energy and power-efficient consumption of desktop computer systems by using the NComputing device. We also propose new approaches to estimate the reliability of k-out-of-n systems based on the delta method. The k-out-of-n system consisting of n subsystems works if and only if at least k-of-the-n subsystems work. More specificly, we develop approaches to obtain the reliability estimation for the k-out-of-n systems which is composed of n independent and identically distributed subsystems where each subsystem （or energy-efficient usage application） can be assumed to follow a two-parameter exponential lifetime distribution function. The detailed derivations of reliability estimation of k-out-of-n systems based on the biased-corrected estimator, known as delta method, the uniformly minimum variance unbiased estimate （UMVUE） and maximum likelihood estimate （MLE） are discussed. An energy-management NComputing application is discussed to illustrate the reliability results in terms of the energy consumption usages of a computer system with qua（t-core, 8 GB of RAM, and a GeForce 9800GX-2 graphics card to perform various complex applications. The estimated reliability values of systems based on the UMVUE and the delta method differ only slightly. Often the UMVUE of reliability for a complex system is a lot more difficult to obtain, if not impossible. The delta method seems to be a simple and better approach to obtain the reliability estimation of complex systems. The results of this study also show that, in practice, the NComputing architecture improves both energy cost saving and energy efficient living spaces.展开更多
In present paper, the properties of the Bayes Shrinkage estimator is studied for the measure of dispersion of an inverse Gaussian model under the Minimax estimation criteria.
THROUGHOUT this note, the following notations are used: For a matrix A, A】0 means thatA is positive definite symmetric; R （A）, A′and A<sup>-</sup> stand for the column space, transposeand any g-inverse...THROUGHOUT this note, the following notations are used: For a matrix A, A】0 means thatA is positive definite symmetric; R （A）, A′and A<sup>-</sup> stand for the column space, transposeand any g-inverse of A respectively; P<sub>A</sub> = A （A′A）<sup>-</sup> A′and P<sub>A</sub> = I<sub>k</sub> - P<sub>A</sub>, where I<sub>k</sub> is theidentity matrix of order k that is the number of rows in A. R<sup>m×n</sup> is the totality of m×n realmatrices.展开更多
基金Supported by the National Natural Science Foundation of China.
文摘For a seemingly Unrelated regression system with the assumption of normality,a necessary and sufficient condition for the existence of the Uniformly Minimum Risk Unbiased (UMRU)estimator of regression coefficients under strictly convex loss is obtained;it is proved that any unbiased estimator can not improve the least squares estimator;it is also shown that no UMRU estimator exists under missing observations.
基金This work was supported by the National Natural Science Foundation of China (Grant No. 19871088).
文摘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.
文摘Throughout this note, the following notations are used. For matrices A and B,A】B means that A-B is positive definite symmetric, A×B denotes the Kroneckerproduct of A and B R（A）, A’ and A<sup>-</sup> stand for the column space, the transpose andany g-inverse of A, respectively; P<sub>A</sub>=A（A’A）<sup>-</sup>A’;for s×t matrix B=（b<sub>1</sub>…b<sub>t</sub>）,vec（B） de-notes the st-dimensional vector （b<sub>1</sub>′b<sub>2</sub>′…b<sub>t</sub>′）′, trA stands for the trace of the square ma-trix A.
基金supported by Rutgers CCC Green Computing Initiative
文摘Opting to follow the computing-design philosophy that the best way to reduce power consumption and increase energy efficiency is to reduce waste, we propose an architecture with a very simple ready-implementation by using an NComputing device that can allow multi-users but only one computer is needed. This intuitively can save energy, space as well as cost. In this paper, we propose a simple and realistic NComputing architecture to study the energy and power-efficient consumption of desktop computer systems by using the NComputing device. We also propose new approaches to estimate the reliability of k-out-of-n systems based on the delta method. The k-out-of-n system consisting of n subsystems works if and only if at least k-of-the-n subsystems work. More specificly, we develop approaches to obtain the reliability estimation for the k-out-of-n systems which is composed of n independent and identically distributed subsystems where each subsystem （or energy-efficient usage application） can be assumed to follow a two-parameter exponential lifetime distribution function. The detailed derivations of reliability estimation of k-out-of-n systems based on the biased-corrected estimator, known as delta method, the uniformly minimum variance unbiased estimate （UMVUE） and maximum likelihood estimate （MLE） are discussed. An energy-management NComputing application is discussed to illustrate the reliability results in terms of the energy consumption usages of a computer system with qua（t-core, 8 GB of RAM, and a GeForce 9800GX-2 graphics card to perform various complex applications. The estimated reliability values of systems based on the UMVUE and the delta method differ only slightly. Often the UMVUE of reliability for a complex system is a lot more difficult to obtain, if not impossible. The delta method seems to be a simple and better approach to obtain the reliability estimation of complex systems. The results of this study also show that, in practice, the NComputing architecture improves both energy cost saving and energy efficient living spaces.
文摘In present paper, the properties of the Bayes Shrinkage estimator is studied for the measure of dispersion of an inverse Gaussian model under the Minimax estimation criteria.
文摘THROUGHOUT this note, the following notations are used: For a matrix A, A】0 means thatA is positive definite symmetric; R （A）, A′and A<sup>-</sup> stand for the column space, transposeand any g-inverse of A respectively; P<sub>A</sub> = A （A′A）<sup>-</sup> A′and P<sub>A</sub> = I<sub>k</sub> - P<sub>A</sub>, where I<sub>k</sub> is theidentity matrix of order k that is the number of rows in A. R<sup>m×n</sup> is the totality of m×n realmatrices.
