search

johannesulf / nautilus

Neural Network-Boosted Importance Nested Sampling for Bayesian Statistics
https://nautilus-sampler.readthedocs.io
MIT License
73 stars 8 forks source link

evidence #39

Closed Jane550 closed 1 year ago

Jane550 commented 1 year ago

Does 'sampler.evidence()' return the Bayesian evidence of the model defined as eqn.(17) in https://ned.ipac.caltech.edu/level5/Sept13/Trotta/Trotta4.html? Can it be directly used to calculate the Bayesian ratio for model comparison?

johannesulf commented 1 year ago

Hi @Jane550, if I understand the article you linked to correctly, sampler.evidence() returns the (log of the) Bayesian evidence as defined in equation (17).

johannesulf commented 1 year ago

And yes, you should be able to use it to calculate the Bayesian evidence ratio for model comparison.

Jane550 commented 1 year ago

Thanks for your reply!