A linear regression model, when a design matrix has not full column rank and a covariance matrix is singular, is considered. The problem of testing hypotheses on mean value parameters is studied. Conditions when a hypothesis can be tested or when need not be tested are given. Explicit forms of test statistics based on residual sums of squares are presented.
@article{bwmeta1.element.bwnjournal-article-doi-10_7151_dmps_1078,
author = {Eva Fi\v serov\'a},
title = {Testing hypotheses in universal models},
journal = {Discussiones Mathematicae Probability and Statistics},
volume = {26},
year = {2006},
pages = {127-149},
zbl = {1128.62069},
language = {en},
url = {http://dml.mathdoc.fr/item/bwmeta1.element.bwnjournal-article-doi-10_7151_dmps_1078}
}
Eva Fišerová. Testing hypotheses in universal models. Discussiones Mathematicae Probability and Statistics, Tome 26 (2006) pp. 127-149. http://gdmltest.u-ga.fr/item/bwmeta1.element.bwnjournal-article-doi-10_7151_dmps_1078/
[000] [1] L. Kubáček, L. Kubáčková and J. Volaufová, Statistical models with linear structures, Veda, Bratislava 1995.
[001] [2] P.B. Pantnaik, The non-central χ² and F-distribution and their applications, Biometrika 36 (1949), 202-232.
[002] [3] C.R. Rao, Linear statistical inference and its applications, J. Wiley and Sons, New York-London-Sydney 1965. | Zbl 0137.36203
[003] [4] C.R. Rao and S.K. Mitra, Generalized inverse of matrices and its applications, J. Wiley and Sons, New York-London-Sydney-Toronto 1971. | Zbl 0236.15004
[004] [5] J. Ryšavý, Higher geodesy, Česká matice technická, Praha 1947 (in Czech).