On Assessing Distributional Properties of Multidimensional Variables

By H.E.T. Holgersson

July 2003

Goteborg University

ISBN: 91-22-02003-9

114 pages, Illustrated, 6 1/2" x 9 1/2"

$58.50 Paper Original

This Ph.D. thesis concerns the problem of assessing distributional properties in multivariate regression models with possibly related marginal models. Such models are of great importance in most branches of statistics. The four papers in this thesis deal with different distributional aspects of relevance for all models. Paper 1 concerns the problem of assessing normality in the multivariate regression model. Paper 2 concerns the problem of assessing normality in situations when the data is heteroscedastic or autocorrelated. Paper 3 concerns testing for heteroscedasticity in linear regression models. Paper 4 presents four different types of tests for autocorrelation to be used in SUR models or multivariate regressions.

Math; Statistics

Return to Coronet Books main page