CMStatistics 2015: Start Registration
View Submission - CMStatistics
B0323
Title: Adapting the Bernstein copula to modeling cylindrical and spherical data with applications Authors:  Michael Wiper - Universidad Carlos III de Madrid (Spain) [presenting]
Concepcion Ausin - Universidad Carlos III de Madrid (Spain)
Abstract: A nonparametric approach is proposed for estimating the dependence relationships between circular variables and other circular or linear variables using copulas. The proposed method is based on the use of Bernstein copulas which allow for the approximation of any kind of dependence structure, including asymmetric relationships. We present a simple procedure to adapt Bernstein copulas to the circular framework and guarantee that the constructed bivariate distributions are strictly continuous. Applications to wind direction data are provided.