CosmoCoffee

[Sunny Vagnozzi]

Hi everyone,

I used getdist to analyze some chains (not produced by CosmoMC), which are in the usual format weight -log(like) param_1 param_2 (I have only 2 parameters).

Despite getdist worked properly (convergence properly estimated, scripts for plots produced), it produced neither a .margestats file, nor a .likestats one (it did produce .covmat, .corr, and .converge ones). Any idea why this might be happening?

Cheers,
Sunny

[Luke Hart]

Can you show us your output to terminal from running getdist??

[Sunny Vagnozzi]

Hi Luke, thanks I figured out the mistake myself, basically one of the entries in the distparams.ini contained one parameter I was not using, e.g. something like params_2D = param_1 param_2 param_3, when I was not varying param_3 in the chains. Hope this can help someone who incurs in the same mistake!