| 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 |
|
| Crossref DOI |
|
| 関連DOI |
|
|
|
| 関連URI |
|
|
|
| 関連情報 |
|
|
|
| 概要 |
The analysis of repeated-measures data is studied in the situation with some suspicious data which deviate relatively from other data and which might be generated by different mechanisms. If we assume... the alternative error distributions which are heavy-tailed relative to the normal distribution, the estimates might be relatively unaffected by outliers. From this point of view, we assume scale mixtures of multivariate normal distributions as the error distributions, in order to reduce the influence of outliers. The case with missing observations is jointly considered and the method for computing the maximum likelihood estimates is given by applying the EM algorithm. Model selection and detection of outliers are also discussed with real data.続きを見る
|
Mendeley出力
