A test for variance-covarianch parameters in normal linear models

From MaRDI portal





The variance-covariance parameters of the usual linear model are first set as coefficients in a derived linear model and then estimated under the usual Gaussian set up. A Wald type test statistic for these parameters, called the \(K^ 2\)- statistic is then derived and is shown to be the sum of the squares of independent standardized chi-square variables. The statistic is also a quadratic function of the MINQUE estimators of the parameters.











This page was built for publication: A test for variance-covarianch parameters in normal linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3711533)