I'm trying to get a sample from a multivariate distribution which is constructed by copula. this is the steps that i go,is it true? at first i estimate the copula and the marginals then get a sample from my copula and since $x=F^{-1}(u)$ i put the samples(u) in quantile of marginals for example qlnorm(u). is it true?

  • $\begingroup$ If you are able to simulate a mutli-variate uniform vector $U$ with the copula distribution under consideration, then your approach is correct. I think, the simulating $U$ itself is the difficult part. However, if you have access to the correlation matrix, then you can use Cholesky's decomposition to get $X$ - Note that this alternate approach only works for the case where $X$ is multivariate normal $\endgroup$ – kasa Dec 17 '18 at 16:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.