摘要

Let , be two continuous random variables. Investigating the regression dependence of on , respectively, of on , we show that the two of them can have almost opposite behavior. Indeed, given any , we construct a bivariate random vector such that the respective regression dependence measures introduced in Dette et al. (Scand. J. Stat. 40(1):21-41, 2013) satisfy as well as r(1/2)(X,Y) < epsilon.

  • 出版日期2015-11