Minimax estimation of the mean matrix of the matrix-variate normal distribution

Autor

  • Saralees Nadarajah
  • S. Zinodiny
  • S. Rezaei

Słowa kluczowe:

Empirical Bayes estimation, matrix-variate normal distribution, mean matrix, minimax estimation

Abstrakt

<p style="text-align: justify;">In this paper, the problem of estimating the mean matrix &Theta; of a matrix-variate normal distribution with the covariance matrix <em>V I<sub>m</sub></em> is considered under the loss functions,<br /> <em>&omega; tr&delta;-X'Q&delta;-X+1-&omega;tr&delta;-&Theta;'Q&delta;-&Theta; and k[1-e<sup>-tr&delta;-&Theta;'&Gamma;^-1&delta;-&Theta;</sup>]. </em>We construct a class of empirical Bayes estimators which are better than the maximum likelihood estimator under the first loss function for <em>m &gt; p + 1 </em>and hence show that the maximum likelihood estimator is inadmissible. For the case<em> Q = V = Ip</em>, we find a general class of minimax estimators. Also we give a class of estimators that improve on the maximum likelihood estimator under the second loss function for <em>m &gt; p + 1</em> and hence show that the maximum likelihood estimator is inadmissible.</p>

Pobrania

Opublikowane

2016-09-02

Numer

Dział

Artykuły [1035]