Distributions.jl
Distributions.jl copied to clipboard
JuliaStats
A Julia package for probability distributions and associated functions.
The moment generating function of NegativeBinomial is wrong: ```julia julia> using Revise, Distributions, ForwardDiff, Test julia> d = NegativeBinomial(4, 0.2) NegativeBinomial{Float64}(r=4.0, p=0.2) julia> @test ForwardDiff.derivative(Base.Fix1(mgf, d), 0) ≈ mean(d) Test...
__TL;DR__: given the same mean vector and covariance matrix, NumPy can sample from a multivariate normal with these parameters, but `Distributions.MultivariateNormal` says: `PosDefException: matrix is not positive definite; Cholesky factorization...
There's no documentation for `DirichletMultinomial`, and it's also not clear to me what `α0` is supposed to be--the usual parametrization of a `DirichletMultinomial` is just in terms of `n` and...
This is a feature request to implement the following methods: ```julia loglikelihood(d::Normal, ss::NormalStats) loglikelihood(d::AbstractMvNormal, ss::MvNormalStats) ``` This would be useful because many algorithms that use some iterative procedure also need...
As discussed in [#1586](https://github.com/JuliaStats/Distributions.jl/pull/1586), this PR adds a new constructor for the LogNormal distribution that takes logB-mean \mu and scale \sigma in an arbitrary base B and converts it to...
The current docstrint for MvNormal starts like this: ``` help?> MvNormal search: MvNormal MvNormalCanon MvNormalKnownCov MvLogNormal AbstractMvNormal MvNormal Generally, users don't have to worry about these internal details. We provide...
Or should I just use lognormal instead, since the only difference seems to be the length of the tail?
Discussed elsewhere, opening an issue to track it. Since they have a slightly different interface and could benefit from different optimizations, @andreasnoack has suggested we separate multivariate distributions (and probably...
Metadata
Owner
Metadata
A Julia package for probability distributions and associated functions.