GetDist: a Python package for analysing Monte Carlo samples
Abstract
Monte Carlo techniques, including MCMC and other methods, are widely used and generate sets of samples from a parameter space of interest that can be used to infer or plot quantities of interest. This note outlines methods used the Python GetDist package to calculate marginalized one and two dimensional densities using Kernel Density Estimation (KDE). Many Monte Carlo methods produce correlated and/or weighted samples, for example produced by MCMC, nested, or importance sampling, and there can be hard boundary priors. GetDist's baseline method consists of applying a linear boundary kernel, and then using multiplicative bias correction. The smoothing bandwidth is selected automatically following Botev et al., based on a mixture of heuristics and optimization results using the expected scaling with an effective number of samples (defined to account for MCMC correlations and weights). Twodimensional KDE use an automaticallydetermined elliptical Gaussian kernel for correlated distributions. The package includes tools for producing a variety of publicationquality figures using a simple namedparameter interface, as well as a graphical user interface that can be used for interactive exploration. It can also calculate convergence diagnostics, produce tables of limits, and output in latex.
 Publication:

arXiv eprints
 Pub Date:
 October 2019
 arXiv:
 arXiv:1910.13970
 Bibcode:
 2019arXiv191013970L
 Keywords:

 Astrophysics  Instrumentation and Methods for Astrophysics;
 Astrophysics  Cosmology and Nongalactic Astrophysics;
 Physics  Data Analysis;
 Statistics and Probability
 EPrint:
 GetDist 1.0 now released, see https://getdist.readthedocs.io