KernelDensity.jl
KernelDensity.jl copied to clipboard
Give UnivariateKDE distribution-like overloads
Upstreams https://github.com/SciML/DiffEqUncertainty.jl/pull/41 . Allows for it to be "distribution-like". Still missing the random sampling.
Any status update on getting this PR merged?
Is it actually guaranteed that the the support is zero outside the extrema of x
? I would have to look at the code, but I don't think it holds in general. Eg for a Normal
kernel (which is the default), the support is the real line. x
are just the points where it is precalculated.