DOI: 10.12732/ijam.v37i2.9
RANDOM CORRELATION MATRICES
AND ONE DECOMPOSITION OF THE
WISHART DISTRIBUTION
Anna Nikolova 1,§ , Tzvetan Ignatov 2 , Bozhidar Dyakov 3
1 Technical University of Varna
Department of Mathematics and Physics
Varna-9010, BULGARIA
2 Sofia University “St. Kliment Ohridski”
Faculty of Economics and Business Administration
Sofia-1113, BULGARIA
3 Technical University of Varna
Department of Navigation
Varna-9010, BULGARIA
Abstract. In the proposed article, random matrices whose distribution coincides with the Wishart distribution are considered. Their elements, which are dependent, are represented as algebraic functions of independent random variables. The densities of the independent random variables are also indicated. A Wishart matrix construction is thus obtained. The present paper shows the relationship between the proposed Wishart matrix construction and correlations and partial correlations. The considered construction was also used to obtain a factorization of the determinant of a correlation matrix.
How
to cite this paper?
DOI: 10.12732/ijam.v37i2.9
Source: International Journal of Applied Mathematics
ISSN printed version: 1311-1728
ISSN on-line version: 1314-8060
Year: 2024
Volume: 37
Issue: 2
References
[1] T.W. Anderson, An Introduction to Multivariate Statistical Analysis, 3rd ed., Wiley, New York (2003).
[2] J. Banfield, A. Raftery, Model-based Gaussian and non-Gaussian clustering, Biometrics, 49 (1993), 803-821.
[3] J. Barnard, R. McCulloch, Xiao-Li Meng, Modeling covariance matrices in terms of standard deviations and correlations, with application to shrink age, Statistica Sinica, 10 (2000), 1281-1311.
[4] H. Bensmail, G. Celeux, A. Raftery, C. Robert, Inference in model-based cluster analysis, Statist. Comput., 7 (1997), 1-10.
[5] G.E.P. Box, G.M. Jenkins and G. Reinsel, Time Series Analysis-Forecasting and Control, Revised 3rd ed., Prentice Hall, NJ (1994).
[6] G. Celeux, G. Govaert, Gaussian parsimonious clustering models, J. Pattern Recognition Soc., 28 (1995), 781-793.
[7] R. Couillet, Z. Liao, Random Matrix Methods for Machine Learning, Cambridge University Press (2022).
[8] N.A.C. Cressie, Statistics for Spatial Data, Revised ed., Wiley, New York (1993).
[9] N. Giri, Multivariate Statistical Analysis, Marcel Dekker, New York (2004).
[10] Tz. Ignatov, A. Nikolova, About Wishart’s distribution, Annuaire de l’Universite de Sofia “St. Kliment Ohridski”, Faculty of Economics and Business Administration, 3 (2004), 79-94.
[11] Tz. Ignatov, A. Nikolova, A decomposition of the Wishart distribution, Annuaire de l’Universite de Sofia “St. Kliment Ohridski”, Faculty of Economics and Business Administration, 4 (2005), 131-147.
[12] H. Joe, Generating random correlation matrices based on partial correlations, J. Multivariate Anal., 97 (2006), 2177-2189.
[13] D. Kurowicka, R. Cooke, Completion problem with partial correlation vines, Linear Algebra Appl., 418, No 1 (2006), 188-200.
[14] A. Nikolova, Evaluation the Moments of the Wishart Distribution, Adv. Stud. Contemp. Math., 15, No 2 (2007), 171-186.
[15] J. Pinheiro, D. Bates, Unconstrained parametrizations for variance covariance matrices, Statist. Comput., 6 (1996), 289-296.
[16] M. Pourahmadi, Joint mean-covariance models with applications to longitudinal data: Unconstrained parameterization, Biometrika, 86 (1999), 677-690.
[17] M. Pourahmadi, Maximum likelihood estimation of generalized linear models for multivariate normal covariance matrix, Biometrika, 87 (1999), 425-435.
[18] M. Pourahmadi, Covariance estimation: The GLM and regularization perspectives. Statist. Sci., 26, No 3 (2011), 369-387.
[19] E. Veleva, Joint densities of correlation coefficients for samples from multivariate
standard normal distribution, Pliska Stud. Math. Bulgar., 18 (2007), 379-386.
(c) 2010-2024, Academic Publications, Ltd.; https://www.diogenes.bg/ijam/