Suppose I have a 3-Copula which I would like to construct with two 2-Copula's, as the following construct:

$$ C_2(u, C_1(v,w)) = C(u,v,w) $$

My question is, if $C$ is known to be a valid copula (i.e. follows the properties such as groundedness, etc ...), then are $C_1$ and $C_2$ unique? I think this is the reverse way of looking at the copula compatibility problem, but I am not 100% sure.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.