Rao, C. R. ; Ali, Hydar (1997) An overall test for multivariate normality Student, 2 . pp. 317324. ISSN 00392685

PDF
 Author Version
400kB 
Official URL: http://www.dtic.mil/cgibin/GetTRDoc?AD=ADA332223&...
Abstract
There are a number of methods in the statistical literature for testing whether observed data came from a multivariate normal(MVN) distribution with an unknown mean vector and covariance matrix. Let X_{1}, ...,X_{n} be an iid sample of size n from a pvariate normal distribution. Denote the sample mean and sample variancecovariance matrix by X̅ and S respectively. Most of the tests of multivariate normality are based on the results that Y_{i}=S^{½}(X_{i}X̅), i=1,..., n, are asymptotically iid as pvariate normal than zero mean vector and identity covariance matrix. Tests developed by Andrews et al., Mardina and others are direct functions of Y_{i}. We note that the N=np components of the Y_{i}'s put together can be considered as an asymptotically iid sample of size N from a univariate normal any well known test based on N independent observations for univariate normality. In Particular we can use univariate skewness and kurtosis tests, which are sensitive to deviations from normality.
Item Type:  Article 

Source:  Copyright of this article belongs to Sunday School Board of the Southern Baptist Convention. 
ID Code:  71898 
Deposited On:  28 Nov 2011 04:20 
Last Modified:  18 May 2016 17:23 
Repository Staff Only: item control page