|
Organizers |
Inference and Hypothesis Testing in Mixed Linear Models
by
Miguel Fonseca,
New University of Lisbon
Coauthors: Joao Tiago Mexia & Roman Zmyslony
Using commutative Jordan algebras and their principal basis, it is pos sible to derive UMVUE estimators for estimable vectors and variance com ponents in mixed linear models using the model's canonical form. When certain conditions are met it is possible to use usual F tests, but when this does not happen generalized F tests must be used. An application of these results to cross-nested models is shown.
Date received: July 22, 2007
Copyright © 2007 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # caur-80.