Optimal discrimination design for copula models

• DOI: 10.48550/arxiv.1601.07739
• Publication Date: 2016-01-01
• Authors: Elisa Perrone; Andreas Rappold; Werner G. Müller
• Keywords: Methodology (stat.ME); FOS: Computer and information sciences; 62K05; 0502 economics and business; 05 social sciences; Statistics - Methodology

Optimum experimental design theory has recently been extended for parameter estimation in copula models. However, the choice of correct dependence structure still requires wider analyses. In this work issue selection is treated by using discrimination techniques. The new proposed approach consists use $D_s$-optimality following an extension corresponding equivalence theory. We also present some examples and highlight strength such a criterion as way to discriminate between various classes dependences.