varying all params

Use of Healpix, camb, CLASS, cosmomc, compilers, etc.
Post Reply
Dana García
Posts: 1
Joined: January 26 2016
Affiliation: Universidad de Colima

varying all params

Post by Dana García » March 08 2016

Hi. I'm a new user of COSMOMC.
I have not modified anything in the code and I run cosmomc varying all params: w, wa, mnu. omegak, nnu,nrun, r, Alens, yhe alpha1, meffsterile. But it's been running for a month approximately and still not over. Anyone know how much time it will take? or if it is necessary to reduce the interval. Also, I get warnings that say:

Warning: Rombint failed to converge;
integral, erro, tol: 10378.4046316585 −1.130727759002248E−005
1.0000000000000000E−007

Antony Lewis
Posts: 1547
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: varying all params

Post by Antony Lewis » March 08 2016

I'm not sure that's a very sensible thing to do, things like w, wa, omegak are very degenerate (esp. if you haven't included supernovae and BAO).
It should be OK in a couple of days if it's going to work at all.
Most likely your initial proposal/scatter widths are not appropriate (given presumably you don't have a covariance matrix for this case and parameters will be very degenerate), or the system has too many wide-open degeneracies.

Post Reply