Open Access
Translator Disclaimer
2016 Measuring multivariate association and beyond
Julie Josse, Susan Holmes
Statist. Surv. 10: 132-167 (2016). DOI: 10.1214/16-SS116


Simple correlation coefficients between two variables have been generalized to measure association between two matrices in many ways. Coefficients such as the RV coefficient, the distance covariance (dCov) coefficient and kernel based coefficients are being used by different research communities. Scientists use these coefficients to test whether two random vectors are linked. Once it has been ascertained that there is such association through testing, then a next step, often ignored, is to explore and uncover the association’s underlying patterns.

This article provides a survey of various measures of dependence between random vectors and tests of independence and emphasizes the connections and differences between the various approaches. After providing definitions of the coefficients and associated tests, we present the recent improvements that enhance their statistical properties and ease of interpretation. We summarize multi-table approaches and provide scenarii where the indices can provide useful summaries of heterogeneous multi-block data. We illustrate these different strategies on several examples of real data and suggest directions for future research.


Download Citation

Julie Josse. Susan Holmes. "Measuring multivariate association and beyond." Statist. Surv. 10 132 - 167, 2016.


Received: 1 December 2015; Published: 2016
First available in Project Euclid: 17 November 2016

zbMATH: 1359.62212
MathSciNet: MR3573303
Digital Object Identifier: 10.1214/16-SS116

Keywords: $k$ nearest-neighbor graph , dCov coefficient , distance matrix , HHG test , measures of association between matrices , multi-block data analyses , permutation tests , RV coefficient , tests of independence

Rights: Copyright © 2016 The author, under a Creative Commons Attribution License


Vol.10 • 2016
Back to Top