Recognizing and visualizing copulas : an approach using local Gaussian approximation

摘要: Abstract In this paper we examine the relationship between a newly developed local dependence measure, Gaussian correlation, and standard copula theory. We are able to describe characteristics of structure in different models terms correlation. Further, construct goodness-of-fit test for bivariate models. An essential ingredient is use canonical correlation pseudo-observations which make independent margins, so that it genuine structure. A Monte Carlo study reveals performs very well compared commonly used alternative test. also propose two types diagnostic plots can be investigate cause rejected null. Finally, our methods applied “classical” insurance data set.