<abstract xmlns="http://www.w3.org/1999/xhtml"><p> The paper is focused on decomposition of a multivariate model into a system of two simpler models. The multiresponses are considered to be in- dependent with the same covariance matrix. Tests are proposed to identify which of the two models should be used in order to obtain more efficient estimators. In a case of partly known model parameters, a tolerance domain for negligible parameters is given.</p></abstract>

The paper is focused on decomposition of a multivariate model into a system of two simpler models. The multiresponses are considered to be in- dependent with the same covariance matrix. Tests are proposed to identify which of the two models should be used in order to obtain more efficient estimators. In a case of partly known model parameters, a tolerance domain for negligible parameters is given.

Decomposition of multivariate statistical models

Department of Mathematical Analysis and Applications of Mathematics Faculty of Science Palack´y University 17. listopadu 12 CZ–771-46 Olomouc CZECH REPUBLIC

Tatra Mountains Mathematical Publications

{"article-title":"Decomposition of multivariate statistical models"}

The paper is focused on decomposition of a multivariate model into a system of two simpler models. The multiresponses are considered to be in- dependent with...

Decomposition of multivariate statistical models

Publicado en línea: 13 nov 2012

Páginas: 33 - 43

DOI: https://doi.org/10.2478/v10127-012-0004-1

Palabras clave
multivariate model, decomposition of model, unbiased estimator, test for decomposition, tolerance domain for negligible parameters.

This content is open access.

Decomposition of multivariate statistical models

Publicado en línea: 13 nov 2012

Páginas: 33 - 43

DOI: https://doi.org/10.2478/v10127-012-0004-1

Palabras clavemultivariate model, decomposition of model, unbiased estimator, test for decomposition, tolerance domain for negligible parameters.

This content is open access.

Palabras clave
multivariate model, decomposition of model, unbiased estimator, test for decomposition, tolerance domain for negligible parameters.