CosmoCoffee Forum Index CosmoCoffee

 
 FAQFAQ   SearchSearch  MemberlistSmartFeed   MemberlistMemberlist    RegisterRegister 
   ProfileProfile   Log inLog in 
Arxiv New Filter | Bookmarks & clubs | Arxiv ref/author:

Convergence of chains when a parameter has a one tail distribution
 
Post new topic   Reply to topic    CosmoCoffee Forum Index -> Computers and software
View previous topic :: View next topic  
Author Message
Daniel Francisco Boriero



Joined: 09 Mar 2009
Posts: 16
Affiliation: Bielefeld

PostPosted: December 08 2016  Reply with quote

Dear All,

I am testing a model with exponential parametrization of the reionization and I am facing the problem to run chains on cosmomc with a parameter (the exponent) which has a one tail distribution.

Do you have any advice to help the convergence of the Monte-Carlo? I would like to test a few decades of this parameter, but the distribution becomes flat quickly and it is taking forever to reach convergence.


Thanks a lot,
Daniel
Back to top
View user's profile [ Hidden ]
Antony Lewis



Joined: 23 Sep 2004
Posts: 1219
Affiliation: University of Sussex

PostPosted: December 08 2016  Reply with quote

Re-parameterize, e.g. using the log of that variable as the MCMC parameter? (though note that this also changes the implicit flat prior).
Back to top
View user's profile [ Hidden ] Visit poster's website
Daniel Francisco Boriero



Joined: 09 Mar 2009
Posts: 16
Affiliation: Bielefeld

PostPosted: December 08 2016  Reply with quote

Indeed, the log would work, but since the parametrization chosen has some physical meaning, I would have to fix the prior to transform for a flat prior on the original parameter.


Thanks Antony,
Daniel
Back to top
View user's profile [ Hidden ]
Daniel Francisco Boriero



Joined: 09 Mar 2009
Posts: 16
Affiliation: Bielefeld

PostPosted: December 13 2016  Reply with quote

Dear Antony,

To fix the prior, would you sugest to use the function AdjustPrior(), on GetDist?

Moreover, is it still enough to process the chains only using these lines below?

! coldata(1,i) = coldata(1,i)*exp(-chisq/2)
! coldata(2,i) = coldata(2,i) + chisq/2


Thanks,
Daniel
Back to top
View user's profile [ Hidden ]
Antony Lewis



Joined: 23 Sep 2004
Posts: 1219
Affiliation: University of Sussex

PostPosted: December 13 2016  Reply with quote

I think that should work. (you may also want to adjust the chi2_xxx derived parameters, if you use them for anything)
Back to top
View user's profile [ Hidden ] Visit poster's website
Daniel Francisco Boriero



Joined: 09 Mar 2009
Posts: 16
Affiliation: Bielefeld

PostPosted: December 13 2016  Reply with quote

Ok, I am only interested in the total chi2 anyway. I was concerned that these new chi2_xxx derived parameters had changed the meaning of the first two columns, making the AdjustPrior() deprecated.

Thanks for explaining.


Cheers,
Back to top
View user's profile [ Hidden ]
Display posts from previous:   
Post new topic   Reply to topic    CosmoCoffee Forum Index -> Computers and software All times are GMT + 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group. Sponsored by WordWeb online dictionary and dictionary software.