High-dimensional inference on covariance structures via the extended cross-data-matrix methodology

2015/03/22 by Kazuyoshi Yata, Makoto Aoshima, Yata, Kazuyoshi +1
Mathematics · #62H12 #62H15 #FOS: Computer and information sciences #FOS: Mathematics #Methodology (stat.ME) #Statistics Theory (math.ST) #math.ST #msc:62H12 #msc:62H15 #stat.ME #stat.TH

paper · pdf · doi:10.48550/arxiv.1503.06492

arxiv created 2015/03/22 · arxiv updated 2015/03/24

Abstract

In this paper, we consider testing the correlation coefficient matrix between two subsets of high-dimensional variables. We produce a test statistic by using the extended cross-data-matrix (ECDM) methodology and show the unbiasedness of ECDM estimator. We also show that the ECDM estimator has the consistency property and the asymptotic normality in high-dimensional settings. We propose a test procedure by the ECDM estimator and evaluate its asymptotic size and power theoretically and numerically. We give several applications of the ECDM estimator. Finally, we demonstrate how the test procedure performs in actual data analyses by using a microarray data set.

Related

• Covariance matrix testing in high dimension using random projections
• High-dimensional covariance matrix regularization using informative targets
• Better Together: Cross and Joint Covariances Enhance Signal Detectability in Undersampled Data
• Fast Robust Correlation for High-Dimensional Data
• On testing the equality of high dimensional mean vectors with unequal covariance matrices