Testing for non-normality in multivariate regression with nonspherical disturbances

dc.contributor.author	Holgersson, Thomas
dc.date.accessioned	2011-02-10T11:56:17Z
dc.date.available	2011-02-10T11:56:17Z
dc.date.issued	2002-09-01
dc.identifier.issn	0349-8034
dc.identifier.uri	http://hdl.handle.net/2077/24417
dc.description.abstract	Statistical diagnostic testing is often associated with erratic conclusions due to the fact that a test against one certain specification may be highly sensitive to another specification. This paper concerns assessing normality of autocorrelated or heteroscedastic variables. It is shown why the type I error of skewnesslkurtosis test limits 100% if the data are not i.i.d. We propose a set of tests for non-normality, which are robust to autocorrelationiheteroscedasticity, covering a wide class of situations. The size and power of the tests are investigated by Monte Carlo techniques.	sv
dc.format.extent	37	sv
dc.language.iso	eng	sv
dc.publisher	University of Gothenburg	sv
dc.relation.ispartofseries	Research Report	sv
dc.relation.ispartofseries	2002:9	sv
dc.subject	tests of non-normality	sv
dc.subject	multivariate analysis	sv
dc.subject	heteroscedasticity	sv
dc.subject	autocorrelation	sv
dc.title	Testing for non-normality in multivariate regression with nonspherical disturbances	sv
dc.type	Text	sv
dc.type.svep	report	sv