| 作成者 |
|
|
|
| 本文言語 |
|
| 出版者 |
|
|
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 号 |
|
| 開始ページ |
|
| 終了ページ |
|
| 出版タイプ |
|
| アクセス権 |
|
| JaLC DOI |
|
| 関連DOI |
|
|
|
|
|
|
|
| 関連URI |
|
|
|
|
|
|
|
| 関連情報 |
|
|
|
|
|
|
|
| 概要 |
Let π_i be a normal population N (μ_i, σ^2) with an unknown mean μ_i and a common known variance σ^2 for i=1,2,3. We consider the problem of estimating μ^^*=max(μ_1, μ_2, μ_3), based on the sample me...ans X, Y, Z drawn from π_i for i=1,2,3. This paper presents two estimators for μ^^*, namely, the Pitman estimator μ^^^_p, which has never been derived and suspected by Blumenthal (1984) to be very complex so that Monte Carlo simulation is only a possible way to investigate its performances, and the estimator μ^^^ , which is derived by considering the class of linear combination of X, Y, and Z. Numerical comparisons of μ^^^_p, μ^^^ and an ordinal estimator μ^^^_m=max(X, Y, Z) are made on the criteria of the bias (BIAS) and mean square error (MSE).続きを見る
|
Mendeley出力
