Revista Integración, temas de matemáticas.
Vol. 22 No. 1 y 2 (2004): Revista Integración, temas de matemáticas
Research and Innovation Articles

Non-null Distribution of The Likelihood Ratio Statistic for Testing Multisample Compound Symmetry

María E. Castañeda
Bio
Daya K. Nagar
Bio

Published 2004-09-16

Keywords

  • Compound symmetry,
  • distribution; likelihood ratio test,
  • intraclass correlation,
  • null moments

How to Cite

Castañeda, M. E., & Nagar, D. K. (2004). Non-null Distribution of The Likelihood Ratio Statistic for Testing Multisample Compound Symmetry. Revista Integración, Temas De matemáticas, 22(1 y 2), 59–66. Retrieved from https://revistas.uis.edu.co/index.php/revistaintegracion/article/view/512

Abstract

The non-null distribution of the modified likelihood ratio test statistic Λ∗ for testing multisample compound symmetry of q multivariate Gaussian models is derived. The non-null moments of Λ∗ are obtained in terms of Lauricella's hypergeometric function. The non-null distribution is expressed in terms of H-function. 

Downloads

Download data is not yet available.

References

[1]T. W. Anderson.An Introduction to Multivariate Statistical Analysis, 2ndEdition, John Wiley & Sons, New York, 1984.

[2]A. K. GuptaandD. K. Nagar. “Nonnull distribution of the likelihood ratiocriterion for testing equality of covariance matrices under intraclass correlationmodel”.Communications in Statistics-Theory and Methods,16(1987), no. 11,3323–3341.

[3]A. K. GuptaandD. K. Nagar.Matrix Variate Distributions, Chapman &Hall/CRC, Boca Raton, 2000.

[4]A. M. MathaiandR. K. Saxena.TheH-Function with Applications inStatistics and Other Disciplines, Halsted Press [John Wiley & Sons], New York-London-Sidney, 1978.

[5]D. K. Nagar,S. K. JainandA. K. Gupta. “On testing circular station-arity and related models”.Journal of Statistical Computation and Simulation,29(1988), no. 3, 225–239.

[6]D. K. NagarandM. E. Castañeda. “Testing multisample compound sym-metry”.Advances and Applications in Statistics,3(2003), no. 3, 285–292.

[7] D. K. NagarandC. C. Sánchez. “Exact percentage points for testingmultisample sphericity in the bivariate case”.Communications in Statistics-Simulation and Computation,33(2004), no. 2, 447–457.

[8]D. K. Nagar, J. ChenandA. K. Gupta. “Distribution and percentage pointsof the likelihood ratio statistic for testing circular symmetry”.ComputationalStatistics and Data Analysis,47(2004), no. 1, 79–89.

[9]D. K. Nagar, E. ZarrazolaandE. Bedoya. “Percentage points for testingintraclass correlation structure”.Journal of Probability and Statistical Science,2(2004), no. 2, 199–211.

[10]I. OlkinandS. J. Press. “Testing and estimation for a circular stationarymodel”.Annals of Mathematical Statistics,40(1969), 1358–1373.

[11]H. M. SrivastavaandP. W. Karlsson.Multiple Gaussian HypergeometricSeries, John Wiley & Sons, New York, 1985.

[12]David F. Votaw. “Testing compound symmetry in a normal multivariate dis-tribution”.Annals of Mathematical Statistics,19(1948), 447-473.

[13]S. S. Wilks. “Sample criteria for testing equality of means, equality of variancesand equality of covariances in a normal multivariate distribution”.Annals ofMathematical Statistics,17(1946), 257–281.