Abstract
The generalized linear model is considered in the multidimensional ease; the consistency and the asymptotic normality of the M.L. estimator are proved; the problem of the estimation of the unknown parameter under linear restraint is investigated; then it is possible to justify the test of a linear hypothesis by the Wald test the, L. R. test and the Lagrange multiplier test, the statistic of which are asymptotically distributed according the χ2 distribution. Finally the properties of the separability of hypotheses are extended to this model. An example in muiti-variate probit analysis is given.

This publication has 7 references indexed in Scilit: