Asymptotic Mean and Variance of Gini Correlation Under Contaminated Gaussian Model

This paper establishes the asymptotic closed forms of the expectation and variance of the Gini correlation (GC) under a particular type of bivariate contaminated Gaussian model emulating a frequently encountered scenario in statistical signal processing. Monte Carlo simulation results verify the correctness of the theoretical results established in this paper. In order to gain further insight into GC, we also compare GC to Pearson's product moment correlation coefficient, Kendall's tau, and Spearman's rho by means of root mean squared error. The newly explored theoretical and simulational findings not only deepen the understanding of the rather new GC, but also shed new light on the topic of correlation theory, which is widely applied in statistical signal processing.