Skip to content

exord/bayev

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

41 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Bayev

A package to estimate Bayesian marginal likelihoods based on posterior samples.

The estimators

The package implements two main methods to estimate the marginal likelihood:

In addition, a small module to compute the Harmonic estimator (described, for example, by Kass & Raftery 1995), is also provided.

Usage

The package provides functions to compute the estimators in each of the packages (chib.compute_cj_estimate and perrakis.compute_perrakis_estimate), and a "meta-function" in the run that allow running a series of simulations (run_montecarlo) to estimate the variance of the estimator.

A more detailed documentation is not available (sorry!), but you're welcome to send me questions and comments to rdiaz@unsam.edu.ar.

Disclaimer and attribution

This is work in progress, and some small additions are constantly being implemented. But most of the work was done during my postdoc at the University of Geneva. So, if you use this code, please cite Díaz et al. 2016.

@ARTICLE{2016A&A...585A.134D,
       author = {{D{\'\i}az}, R.~F. and {S{\'e}gransan}, D. and {Udry}, S. and {Lovis}, C. and {Pepe}, F. and {Dumusque}, X. and {Marmier}, M. and {Alonso}, R. and {Benz}, W. and {Bouchy}, F. and {Coffinet}, A. and {Collier Cameron}, A. and {Deleuil}, M. and {Figueira}, P. and {Gillon}, M. and {Lo Curto}, G. and {Mayor}, M. and {Mordasini}, C. and {Motalebi}, F. and {Moutou}, C. and {Pollacco}, D. and {Pompei}, E. and {Queloz}, D. and {Santos}, N. and {Wyttenbach}, A.},
        title = "{The HARPS search for southern extra-solar planets. XXXVIII. Bayesian re-analysis of three systems. New super-Earths, unconfirmed signals, and magnetic cycles}",
      journal = {\aap},
     keywords = {techniques: radial velocities, methods: data analysis, methods: statistical, planetary systems, Astrophysics - Earth and Planetary Astrophysics},
         year = 2016,
        month = jan,
       volume = {585},
          eid = {A134},
        pages = {A134},
          doi = {10.1051/0004-6361/201526729},
archivePrefix = {arXiv},
       eprint = {1510.06446},
 primaryClass = {astro-ph.EP},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2016A&A...585A.134D},
      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

About

Python package to estimate the Bayesian evidence.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages