Oskar Laverny
Oskar Laverny
Small dimensional archimedean copulas (in particular d=2) are really usefull and sometimes have specific samplign schemes taht are much faster than what we are doing right now. We could try...
This still issed a few tests, and inclusion in the general documentation.
This is a first attempt to add a frailty generator: as the williamson generator, it owuld be constructed from univariate distribution, but the distribution would be used as a frailty...
See https://www.stochasticlifestyle.com/semantic-versioning-semver-is-flawed-and-downgrade-ci-is-required-to-fix-it/
AnderGray suggested in the Joss review to add functionality for Plotting / Visualisation. --------- Yes, this is indeed a very good idea. We might borow a few plots from `BivariateCopulas.jl`...
According to what we found in #133, we should modify the current implementation of \tau and \rho so that they return (d,d) matrices on d-variate copulas, containing bivariate kendall and...
This PR tries to switch from TaylorSeries.jl to TaylorDiff.jl because of better performances of the later. But there is still a few broken tests. See https://github.com/JuliaDiff/TaylorDiff.jl/issues/61 for upstream report of...
This new generator shandles outer/inner power families of archimedean generators. A few tests and inclusion in the general documentaiton is still missing.
Adding a mother class for W and M that is a parametric mixture class to contain the Frechet-hoeffding bounds Should resolve #14. This is just a first draft, maybe we...
TaylorDiff is depending on a lot of stuff it should not, which tanks my TTFX. As already noted in https://github.com/JuliaDiff/TaylorDiff.jl/issues/52#issuecomment-1707801903_ , It would be very beneficial to leverage package preferences...