getdist: plotting output

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Tanner Prestegard
Posts: 2
Joined: February 05 2013
Affiliation: University of Wisconsin - Milwaukee

getdist: plotting output

Post by Tanner Prestegard » February 05 2013

Hello,

I am using getdist to process output from MultiNest in order to estimate the 2D marginalized posteriors for a 5-parameter model of the stochastic gravitational-wave background. My question is fairly simple: what is the difference between the getdist output files [file_root]\_2D\_2\_1 and [file_root]\_2D\_2\_1\_likes ? I have looked all over and haven't found any discussion even. It doesn't look like this is covered in the readme, either. Based on the Matlab plotting scripts that are generated, it looks like the "likes" file is used to plot the full surface and the other file is used to calculate the contours specified in the distparams file.

Also, for some reason I seem to be missing some of the outputs of getdist (based on the readme). It isn't producing a .margestats, .likestats, .PCA, or .minmarge file. I don't need those files (at least right now), but I would like to understand why they are missing and how to obtain them, if possible. Thanks in advance for your help.

Note: I am using the January 2012 version of CosmoMC

Leonidas Christodoulou
Posts: 2
Joined: January 30 2008
Affiliation: University of Sussex

getdist: plotting output

Post by Leonidas Christodoulou » February 08 2013

Hi,

[file_root]_2D_2_1 gives the marginalized posterior, whereas [file_root]_2D_2_1_likes gives the mean likelihoods, see http://cosmologist.info/cosmomc/readme.html.

My understanding is that you should be getting the files you are mentioning, but I'm using an older version of cosmomc, so I can't be sure. Maybe the .ini files has changed?

Tanner Prestegard
Posts: 2
Joined: February 05 2013
Affiliation: University of Wisconsin - Milwaukee

getdist: plotting output

Post by Tanner Prestegard » February 14 2013

Thanks a lot for the response! Very helpful!

Post Reply