| |
 |
| Paper: |
idl_emcee: IDL Implementation of the MCMC Ensemble Sampler |
| Volume: |
538, ADASS XXXII |
| Page: |
385 |
| Authors: |
Ashkbiz Danehkar |
| DOI: |
10.26624/BIVU8108 |
| Abstract: |
Ensemble samplers can be used in statistical analyses of observations
containing instrumental and measurement uncertainties. To facilitate this approach in
the Interactive Data Language (IDL), we present idl_emcee, an implementation of the
affine-invariant Markov chain Monte Carlo (MCMC) ensemble sampler based on the
algorithm prescribed by Goodman & Weare (2010). This IDL library allows us to create
samplers for the probability densities of new data generated by a user-defined function
using prior information, which can be employed to handle uncertainties in empirical
studies of observational data and find confidence levels of new data in computations.
This package has recently been used to propagate uncertainties of measured fluxes into
plasma diagnostics and abundance analyses of ionized gaseous nebulae. It is being built
for IDL, but it has also been tested in the GNU Data Language (GDL), an open-source
IDL compiler that can be deployed on Jupyter notebooks. |
|
|
|
|
|
