0
$\begingroup$

I'm trying to simulate multiple random variables with different taus from the Gumbel copula. For the normal copula it's pretty simple, eg:

library(copula)
myCop <- normalCopula(param=c(0.4,0.2,.3), dim = 3, dispstr = "un")
normtest = rCopula(1000, myCop)

But for the Gumbel copula, it seems to only want to do it is all the taus are the same. I tried:

library(gumbel)
alphas = c(1.5, 1.2, 2)
test = rgumbel(n = 1000, dim = 3, alpha = alphas)

Any ideas?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

It seems that you try to simulate a multivariate Gumbel copula. In multivariate non-elliptical copula, there is a restriction to the parameters. Only one parameter value needs to be fitted to the multivariate data. This is one main drawback for copula function. You need to look for nested copula function, however, it is really hard. So, you may need to use pair-copula construction. Therefore, you can fit Gumbel copula to your data very easily, look for VineCopula or CDVine copula packages in R.

Improve this answer
$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.