A Statistical Procedure for the Latent Profile Model
Bibliographic Data
| ID | 19277223 |
|---|---|
| Authors | YOSHIO TAKANE (0000-0002-4892-2018, Department of Psychology, University of Tokyo, corresponding author) |
| Year | 1976 |
| Volume | 18 |
| Issue | 2 |
| Pages | 82-90 |
| Publication date | 1976-01-01 |
| Open Access | No |
| Type | ARTICLE |
| Venue | Japanese Psychological Research (JOURNAL) |
| Journal identifiers | ISSN: 0021-5368 • E-ISSN: 1468-5884 |
| Publisher | The Japanese Psychological Association (PUBLISHER • JP) |
| DOI | 10.4992/psycholres1954.18.82 |
| OpenAlex | W2525509330 |
| Language | EN |
| References cited | 5 |
A maximum likelihood estimation procedure is presented for the latent profile model when the conditional distributions of manifest variates given specific latent classes are assumed to be normal, together with the associated tests of the goodness of fit. By incorporating means of constraining estimated parameters in various ways, an important class of statistical hypotheses, constant and equality, about the structures of latent classes can be tested. Illustrative examples are included with the results suggesting strongly the usefulness of the current approach
Class (philosophy) · Constant (computer programming) · Econometrics · Goodness of fit · Latent class model · Latent variable · Latent variable model · Local independence · Maximum likelihood · Probabilistic latent semantic analysis · Statistical model · Statistics · Structural equation modeling · Advanced Statistical Methods and Models · Artificial Intelligence · Bayesian Methods and Mixture Models · Computer Science · Mathematics · Statistical Distribution Estimation and Applications
| Highly cited | No |
|---|