Informatics and Applications
2021, Volume 15, Issue 1, pp 72-77
NONASYMPTOTIC ANALYSIS OF BARTLETT-NANDA-PILLAI STATISTIC FOR HIGH-DIMENSIONAL DATA
Abstract
The author gets the computable error bounds for normal approximation of Bartlett-Nanda-Pillai statistic when dimensionality grows proportionally to the sample size. This result enables one to get more precise calculations of the p-values in applications of multivariate analysis. In practice, more and more often, analysts encounter situations when the number of factors is large and comparable with the sample size. The examples include signal processing. The proof is essentially based on the normality of the distribution of the elements of the matrices under consideration with the Wishart distribution. For random variables that are the matrix traces of the product and squares of matrices with the normalized Wishart distribution, convenient upper bounds for 1 - F are found where F is the distribution function of the corresponding matrix trace. Applying the properties of inverse matrices and positive semidefinite matrices, the Bartlett-Nanda-Pillai statistic is bounded from above by a combination of the above-mentioned matrix traces.
[+] References (12)
- Fujikoshi, Y., V. V. Ulyanov, and R. Shimizu. 2010. Multivariate statistics: High-dimensional and large-sample ap-proximations. Hoboken, NJ: John Wiley & Sons. 512 p.
- Johnstone, I. M., and B. Nadler. 2017. Roy's largest root test under rank-one alternatives. Biometrika 104(1):181- 193.
- Akbari, V, S.N. Anfinsen, A. P. Doulgeris, T. Eltoft, G. Moser, and S.B. Serpico. 2016. Polarimetric SAR change detection with the complex Hotelling-Lawley trace statistic. IEEET. Geosci. Remote 54(7):3953-3966.
- Anderson, T. W. 2003. An introduction to multivariate anal-ysis. 3rd ed. Hoboken, NJ: John Wiley & Sons. 742 p.
- Muirhead, R. J. 1970. Asymptotic distributions of some multivariate tests. Ann. Math. Stat. 41(3):1002-1010.
- Lipatiev, A. A., and V. V. Ulyanov. 2017. On computable estimates for accuracy of approximation for the Bartlett- Nanda-Pillai statistic. Siberian Adv. Math. 27(3):153- 159.
- Wakaki, H., Y. Fujikoshi, and V. V. Ulyanov. 2014. Asymptotic expansions of the distributions of MANOVA test statistics when the dimension is large. Hiroshima Math. J. 44(3):247-259.
- Lipatiev, A. A., and V. V. Ulyanov. 2019. Neasimptoticheskiy analiz statistiki Louli-Khotellinga dlya dannykh bol'shoy razmernosti [Nonasymptotic analysis of Lawley- Hotelling statistic for high dimensional data]. Zapiski nauchnykh seminarov POMI [POMI Notes of Scientific Seminars] 486:178-189.
- Shevtsova, I. G. 2011. On the absolute constants in the Berry-Esseen type inequalities for identically distributed summands. arXiv:1111.6554 [math.PR]. Available at: https://arxiv.org/pdf/1111.6554 (accessed December 16, 2020).
- Kawaguchi, Yu., V. V. Ulyanov, and Ya. Fujikoshi. 2010. Priblizheniya dlya statistik, opisyvayushchikh geometricheskie svoystva dannykh bol'shoy razmernosti, s otsenka- mi oshibok [Asymptotic distributions of basic statistics in geometric representation for high-dimensional data and their error bounds]. Informatika i ee Primenen iya - Inform. Appl. 4(1):22-27.
- Ulyanov, V.V., H. Wakaki, and Y. Fujikoshi. 2006. Berry- Esseen bound for high dimensional asymptotic approximation of Wilks' Lambda distribution. Stat. Probabil. Lett. 76(12):1191-1200.
- Coope, I. D. 1949. On matrix trace inequalities and related topics for products of Hermitian matrices. J. Math. Anal.Appl. 188(3):999-1001.
[+] About this article
Title
NONASYMPTOTIC ANALYSIS OF BARTLETT-NANDA-PILLAI STATISTIC FOR HIGH-DIMENSIONAL DATA
Journal
Informatics and Applications
2021, Volume 15, Issue 1, pp 72-77
Cover Date
2021-03-30
DOI
10.14357/19922264210110
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
computable estimates; accuracy of approximation; MANOVA; computable error bounds; Bartlett- Nanda-Pillai statistic; high-dimensional data
Authors
A. A. Lipatiev 
Author Affiliations
Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation
|