On the power of Chatterjee rank correlation
share
Chatterjee (2020) introduced a simple new rank correlation coefficient that has attracted much recent attention. The coefficient has the unusual appeal that it not only estimates a population quantity that is zero if and only if the underlying pair of random variables is independent, but also is asymptotically normal under independence. This paper compares Chatterjee's new coefficient to three established rank correlations that also facilitate consistent tests of independence, namely, Hoeffding's D, Blum-Kiefer-Rosenblatt's R, and Bergsma-Dassios-Yanagimoto's τ^*. We contrast their computational efficiency in light of recent advances, and investigate their power against local alternatives. Our main results show that Chatterjee's coefficient is unfortunately rate sub-optimal compared to D, R, and τ^*. The situation is similar but more subtle for a related earlier estimator of Dette et al. (2013). These results favor D, R, and τ^* over Chatterjee's new coefficient for the purpose of testing independence.
READ FULL TEXT
