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 |
Two types of new latent class model are proposed for questionnaires observed at several time-points, where existence of certain time-dependent latent classes is assumed. The authors concretely derive ...the EM algorithm to obtain maximum likelihood estimates of latent parameters in the respective models, such as mixed proportions and identified states-transition parameters of the respective latent classes. The merit of the algorithm derived here is that the resultant estimates can not be improter solutions. Also, the goodness of fit test is given with some numerical examples. The FORTRAN software is available from the authors.続きを見る
|