Abstract
Understanding and modeling multivariate dependence structures depending upon the direction are challenging but an interest of theoretical and applied researchers. In this article, we introduce a way of looking at directional dependence by using a direction parameter, possibly random in Bayesian setting, expressed as an angle. This construction allows us to model and measure directional dependence in a meaningful way and leads to informative graphical displays. Our focus in this paper will be on the 3-dimensional case.