search

brinckmann / montepython_public

Public repository for the Monte Python Code
MIT License
93 stars 80 forks source link

Sigma value #256

Closed SafaeCosmology closed 1 year ago

SafaeCosmology commented 2 years ago

Hello everyone, I'm trying to run my calculations with MontePython, using the 6 standard parameters plus a new parameter that characterizes my model, when the calculations are finished I don't get the sigma value for my new parameter, but when I run info .... - keep-non-markovian --keep-fraction 0.8 --bins 10 I get sigma, can my result be considered good?

brinckmann commented 2 years ago

Dear Safae,

It is a known bug in MontePython that it sometimes fails to compute credibility intervals when it should be able to, and that this can be fixed by changing the binning (typically to lower values) in the way that you did here, even though that should have no impact on the computation of credibility intervals. If your run is converged (ideally max(R-1) ~ 0.01 or less, or at least 0.02-0.03), and the numbers appear to make sense, then I would say they are reliable despite this workaround.

Best, Thejs

SafaeCosmology commented 2 years ago

Thank you very much, sir.