1. Goodness-of-fit tests for parametric families of Archimedean copulas.
- Author
-
Savu, Cornelia and Trede, Mark
- Subjects
- *
COPULA functions , *GOODNESS-of-fit tests , *DISTRIBUTION (Probability theory) , *DEPENDENCE (Statistics) , *MATHEMATICAL statistics , *RANDOM variables , *MARGINAL distributions , *RANDOM numbers , *STATISTICAL hypothesis testing - Abstract
The article presents information on the finding of goodness-of-fit method in order to test the parametric specification for the function of Archimedean copulas. It informs that copulas are an important tool in multivariate statistics and they capture the pure dependence structure between random variables irrespective of their marginal distributions. Random number generation from Archimedean copulas is stated to be easy because of their simple structure. It is stated that if the dependence structure of the return vector can accurately be described by a parametric Archimedean copula, one could aim at developing credit portfolio models that capture the behavior of joint returns more realistically.
- Published
- 2008
- Full Text
- View/download PDF