Limiting distributions of the likelihood ratio test statistics for independence of normal random vectors. (English) Zbl 1512.62058
Summary: Consider the likelihood ratio test (LRT) statistics for the independence of sub-vectors from a \(p\)-variate normal random vector. We are devoted to deriving the limiting distributions of the LRT statistics based on a random sample of size \(n\). It is well known that the limit is chi-square distribution when the dimension of the data or the number of the parameters are fixed. In a recent work by Y. Qi et al. [Ann. Inst. Stat. Math. 71, No. 4, 911–946 (2019; Zbl 1427.62049)], it was shown that the LRT statistics are asymptotically normal under condition that the lengths of the normal random sub-vectors are relatively balanced if the dimension \(p\) goes to infinity with the sample size \(n\). In this paper, we investigate the limiting distributions of the LRT statistic under general conditions. We find out all types of limiting distributions and obtain the necessary and sufficient conditions for the LRT statistic to converge to a normal distribution when \(p\) goes to infinity. We also investigate the limiting distribution of the adjusted LRT test statistic proposed in [Qi et al., loc. cit.]. Moreover, we present simulation results to compare the performance of classical chi-square approximation, normal and non-normal approximation to the LRT statistics, chi-square approximation to the adjusted test statistic, and some other test statistics.
MSC:
| 62H15 | Hypothesis testing in multivariate analysis |
| 62E20 | Asymptotic distribution theory in statistics |
Keywords:
likelihood ratio test; normal random vector; central limit theorem; chi-square approximation; non-normal limit; high dimension; independenceCitations:
Zbl 1427.62049References:
| [1] | Abramowitz, M.; Stegun, IA, Handbook of mathematical functions (1972), New York: Dover, New York · Zbl 0543.33001 |
| [2] | Bai, Z.; Jiang, D.; Yao, JF; Zheng, S., Corrections to LRT on large-dimensional covariance matrix by RMT, Ann Stat, 37, 6, 3822-3840 (2009) · Zbl 1360.62286 · doi:10.1214/09-AOS694 |
| [3] | Bao, Z.; Hu, J.; Pan, G.; Zhou, W., Test of independence for high-dimensional random vectors based on freeness in block correlation matrices, Electron J Stat, 11, 1, 1527-1548 (2017) · Zbl 1362.62123 · doi:10.1214/17-EJS1259 |
| [4] | Blitzstein JK, Hwang J (2014) Introduction to probability. Chapman & Hall/CRC texts in statistical science. CRC Press, Boca Raton |
| [5] | Bodnar, T.; Dette, H.; Parolya, N., Testing for independence of large dimensional vectors, Ann Stat, 47, 5, 2977-3008 (2019) · Zbl 1436.60018 · doi:10.1214/18-AOS1771 |
| [6] | Chen, SX; Zhang, LX; Zhong, PS, Tests for high-dimensional covariance matrices, J Am Stat Assoc, 105, 490, 810-819 (2010) · Zbl 1321.62086 · doi:10.1198/jasa.2010.tm09560 |
| [7] | Dette, H.; Dörnemann, N., Likelihood ratio tests for many groups in high dimensions, J Multivar Anal, 178, 104605 (2020) · Zbl 1440.62216 · doi:10.1016/j.jmva.2020.104605 |
| [8] | Dörnemann N (2022) Likelihood ratio tests under model misspecification in high dimensions. https://arxiv.org/abs/2203.05423 · Zbl 1520.62064 |
| [9] | Guo W, Qi Y (2021) Asymptotic distributions for likelihood ratio tests for the equality of covariance matrices. Technical report. University of Minnesota Duluth, Department of Mathematics and Statistics. http://arxiv.org/abs/2110.02384 |
| [10] | Jiang, T.; Qi, Y., Likelihood ratio tests for high-dimensional normal distributions, Scand J Stat, 42, 4, 988-1009 (2015) · Zbl 1419.62143 · doi:10.1111/sjos.12147 |
| [11] | Jiang, T.; Yang, F., Central limit theorems for classical likelihood ratio tests for high-dimensional normal distributions, Ann Stat, 41, 4, 2029-2074 (2013) · Zbl 1277.62149 · doi:10.1214/13-AOS1134 |
| [12] | Jiang, D.; Jiang, T.; Yang, F., Likelihood ratio tests for covariance matrices of high-dimensional normal distributions, J Stat Plan Inference, 142, 8, 2241-2256 (2012) · Zbl 1244.62082 · doi:10.1016/j.jspi.2012.02.057 |
| [13] | Jiang, D.; Bai, Z.; Zheng, S., Testing the independence of sets of large-dimensional variables, Sci China Math, 56, 1, 135-147 (2013) · Zbl 1256.62033 · doi:10.1007/s11425-012-4501-0 |
| [14] | Ledoit, O.; Wolf, M., Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size, Ann Stat, 30, 4, 1081-1102 (2002) · Zbl 1029.62049 · doi:10.1214/aos/1031689018 |
| [15] | Li, W.; Chen, J.; Yao, J., Testing the independence of two random vectors where only one dimension is large, Statistics, 51, 1, 141-153 (2017) · Zbl 1369.62123 · doi:10.1080/02331888.2016.1266988 |
| [16] | Muirhead, RJ, Aspects of multivariate statistical theory. Wiley series in probability and mathematical statistics (1982), New York: Wiley, New York · Zbl 0556.62028 |
| [17] | Qi, Y.; Wang, F.; Zhang, L., Likelihood ratio test of independence of components for high-dimensional normal vectors, Ann Inst Stat Math, 71, 4, 911-946 (2019) · Zbl 1427.62049 · doi:10.1007/s10463-018-0666-9 |
| [18] | Schott, JR, Some tests for the equality of covariance matrices, J Stat Plan Inference, 94, 1, 25-36 (2001) · Zbl 0971.62031 · doi:10.1016/S0378-3758(00)00209-3 |
| [19] | Schott, JR, Testing for complete independence in high dimensions, Biometrika, 92, 4, 951-956 (2005) · Zbl 1151.62327 · doi:10.1093/biomet/92.4.951 |
| [20] | Schott, JR, A test for the equality of covariance matrices when the dimension is large relative to the sample sizes, Comput Stat Data Anal, 51, 12, 6535-6542 (2007) · Zbl 1445.62121 · doi:10.1016/j.csda.2007.03.004 |
| [21] | Srivastava, MS; Reid, N., Testing the structure of the covariance matrix with fewer observations than the dimension, J Multivar Anal, 112, C, 156-171 (2012) · Zbl 1273.62128 · doi:10.1016/j.jmva.2012.06.004 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
