Fitting copula with known marginal, Copulas.jl

General Usage
1

An example of fitting a copula using the Copulas.jl package is given as the following in the docs:

MyCop = SurvivalCopula{4,ClaytonCopula,(2,4)}
MyMargs = Tuple{LogNormal,Pareto,Gamma,Normal}
MyD = SklarDist{MyCop, MyMargs}
fitted_model = fit(MyD,data)

Question: Is there a good way to fit the copula when one already has the exact distributions for the marginals, not just the types of distributions? I have fit my marginals to custom distributions that take some time to calculate.

It would be great if MyMargs could be explicitly specified like MyMargs=[Normal(0,1), Gamma(1,1), MyDist(1,1)]

Thanks in advance for the help,

2

If you have a CDF for your marginals, you can just transform to the unit space and fit the Copula there, i.e.,

known_marginals = [Normal(0, 1), Gamma(1, 1), MyDist(1, 1)]
u = cdf.(known_marginals, data)
fit(MyCop, u)