getdist output for 1D plots

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Karsten Jedamzik
Posts: 5
Joined: November 30 2011
Affiliation: University of Montpellier

getdist output for 1D plots

Post by Karsten Jedamzik » April 02 2015

Hi,

A simple question. For a simple chain, the likelihood weighted distribution of a particular variable (e.g. Omegab, p1) should be simply given by the
frequency of this variable occurring in the output .txt files, right ? And that
independent of the first two colums, weight and likelihood.

Now, when I compute the distribution this way from the .txt file myself, I neither find the distribution (with same bin number, no smoothing, etc) as given in the files file_root_p1.likes or file_root_p1.dat (neither of the columns). Though they are close, they are significantly different.

What exactly is given in those files, file_root_p1.likes and file_root_p1.dat ?

Anybody has an idea what I could have done wrong ?

Karsten

Karsten Jedamzik
Posts: 5
Joined: November 30 2011
Affiliation: University of Montpellier

getdist output for 1D plots

Post by Karsten Jedamzik » April 03 2015

Sorry, forget my question, of course I have to weight the entries by the
weight in the first column of the *.txt files, to be correct with MCMC.

Post Reply