Advances and Applications in Statistics
Volume 20, Issue 1, Pages 67 - 88
(January 2011)
|
|
CHARACTERIZATION OF k-VARIATE NORMAL DISTRIBUTION WITH COVARIANCE STRUCTURE, �AND EDF GOODNESS-OF-FIT TESTS
Dhanuja Kasturiratna, Truc T. Nguyen and Arjun K. Gupta
|
Abstract: Characterization of k-variate normal distribution, with covariance matrix under the form �based on UMVU estimator of the density function, is given. This result is used with the transformations given by Rincon-Gallardo et al. [9], in constructing EDF goodness-of-fit tests for testing k‑variate normality with covariance matrix �This transformation and the transformation proposed by Kasturiratna et al. [4] are compared and the powers of the tests are estimated by Monte Carlo method for several alternatives. |
Keywords and phrases: k-variate normal distribution, conditional density function, UMVUE of the density function, characteristic function, differential equation, moments, EDF, power. |
|
Number of Downloads: 357 | Number of Views: 845 |
|