rand5

@lrnv Thank you for fixing that- I can confirm I'm no longer getting the domain errors that were being thrown before. I think this has opened the door to another...

Looking at the Monte Carlo decomposition you described, it seems like the variability between the two approaches (going from X2->X1 versus X1->X2) comes down to whether we evaluate `p1 =...

How about the following for a slightly revised implementation of the generic version: ``` function convolveF(D::SklarDist,Zs::Vector{Float64};N=1000) # The primary should be the function with the largest range MaxRange = [maximum(x->pdf(X,x),Zs)...

Correct, this is just for bivariates. I was imagining that if there were 3 components (A, B and C), then we could do some sort of nested evaluation like `convolveF(convolveF(A,B),C)`...

@lrnv Any interest in collaborating on a technical paper?

With regards to your suggestion to do something like this: ``` function convolve(D::SklarDist{CT, Tuple{T1,T2}}) where {T1

@lrnv Thank you for following up on this. I'm not too familiar with Github-- it looks like #194 was merged as v0.1.30, but I'm running into the following after I...