Abstract
We consider the general linear model with the classical hypothesis. In particular we consider the variance analysis in two factors and we give explicitly the forms of the associated matrixes as well as the statistics of the likelihood ratio test for the null hypothesis. We obtain in a explicit form the distributions of the estimator under hypothesis of normality.
References
T.W. Anderson (1958) An introduction to multivariate statistical analysis. J. Wiley, N.Y.
J.R. Barra (1971) Notions fondamentales de statistique mathématique. Dunod, Paris.
H. Muirhead (1982) Aspects of multivariate statistical theory. J. Wiley, N.Y.
J. Poltronieri (1988) Estudio de formas cuadráticas en el caso multivariado. In: IV Simposio de Métodos Matemáticos Aplicados a las Ciencias Ed. U.C.R.
J. Poltronieri (1988) Formas cuadráticas y formas lineales en estadística multivariada. In: IV Simposio de Métodos Matemáticos Aplicados a las Ciencias Ed. U.C.R.
K. Takeuchi, H. Yanai, B.N. Mukherjee (1984) The foundations of multivariate analysis.Wiley Eastern Limited.
M. Tenenhaus, F. Young (1987) An analysis and synthesis of multiple correspondance analysis, optimal scaling, dual scaling, homogenity analysis and other methods for quantifying categorical multivariate data, Psychometrica 50(1), pp. 91-119.