|
Title
|
|
|
|
The comonotonicity coefficient: a new measure of positive dependence in a multivariate setting
| |
|
Author
|
|
|
|
| |
|
Abstract
|
|
|
|
| In financial and actuarial sciences, knowledge about the dependence structure is of a great importance. Unfortunately this kind of information is often scarce. Many research has already been done in this field e.g. through the theory of comonotonicity. It turned out that a comonotonic dependence structure provides a very useful tool when approxi- mating an unknown but (preferably strongly) positive dependence structure. As a consequence of this evolution, there is a need for a measure which reflects how close a given dependence structure approaches the comonotonic one. In this contribution, we design a measure of (positive) association between n variables (X1,X2, · · · ,Xn) which is useful in this context. The proposed measure, the comonotonicity coefficient (X) takes values in the range [0, 1]. As we want to quantify the degree of comonotonicity, (X) is defined in such a way that it equals 1 in case (X1,X2, · · · ,Xn) is comonotonic and 0 in case (X1,X2, · · · ,Xn) is independent. It should be mentioned that both the marginal distri- butions and the dependence structure of the vector (X1,X2, · · · ,Xn) will have an effect on the resulting value of this comonotonicity coefficient. In a first part, we show how (X) can be designed analytically, by making use of copulas for modeling the dependence structure. In the particular case where n = 2, we compare our measure with the classic dependence measures and find some remarkable relations between our measure and the Pearson and Spearman correlation coefficients. In a second part, we focus on the case of a discounting Gaussian process and we investigate the performance of our comonotonicity coefficient in such an environment. This provides us insight in the reason why the comonotonic structure is a good approximation for the dependence structure. |
| |
|
Language
|
|
|
|
English
| |
|
Source (series)
|
|
|
|
Research paper / UA, Faculty of Applied Economics ; 2006:30
| |
|
Publication
|
|
|
|
Antwerp
:
UA
,
2006
| |
|
Volume/pages
|
|
|
|
42 p.
| |
|
Full text (open access)
|
|
|
|
| |
|