Empirical density contours from an MCMC sample

Specific Domains Probabilistic programming
1

I have a recollection of seeing two-dimensional empirical density contours from an MCMC (Markov-chain Monte Carlo) sample but I didn’t record the information on the package. Was I remembering correctly and, if so, could someone remind me of the package? I am hoping to adapt such code for parametric bootstrap samples of parameter estimates.

2

Maybe the KernelDensityEstimate.jl and KernelDensityEstimatePlotting.jl packages?

The README of KernelDensityEstimate.jl shows what the plots look like.

3

From matplotlib - A plot describing the density of data points in 2D space in Julia - Stack Overflow there is the suggestion

using StatsPlots, KernelDensity
a, b = randn(10000), randn(10000)
dens = kde((a,b))
plot(dens)

which looks pretty nice.

2 Likes