It is shown how to define the canonic formulation for orthogonal models associated to commutative Jordan algebras. This canonic formulation is then used to carry out inference. The case of models with commutative orthogonal block structures is stressed out.
Department of Mathematics, Évora University, Portugal
Bibliografia
[1] D.J. Ferreira, Inducing pivot variables and variance components in orthogonal normal models, Ph.D Thesis 2005.
[2] S.M. Ferreira, Inferęncia para Modelos Ortogonais com Segregação, Tese de Doutoramento 2006.
[3] M. Fonseca, J.T. Mexia and R. Zmyślony, Binary Operation on Jordan algebras and orthogonal normal models, Linear Algebra And Its Applications (2006), 75-86.
[4] A.I. Khuri, Advanced Calculus with Applications in Statistics, 2nd ed., Wiley 2003.
[5] J.T. Mexia, Best linear unbiased estimates, duality of F tests and the Scheffé multiple comparison method in presence of controlled heteroscedasticity, Comp. Stat. Data Analysis 10 (3) (1990).
[7] J.R. Schott, Matrix Analysis for Statistics, Wiley Series in Probability and Statistics 1997.
[8] J. Seely, Linear spaces and unbiased estimators. Application to a mixed linear model, The Annals of Mathenatical Statistics 41 (5) (1970), 1735-1745.
[9] R. Zmyślony, Completeness for a family of normal distributions, Mathematic Statistics, Banach Center Publications 6 (1980), 355-357.