CosmoMC: chi^2 values with action 4 vs action 2

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Jose Lozano
Posts: 19
Joined: June 27 2023
Affiliation: Universidad Nacional Autónoma de México

CosmoMC: chi^2 values with action 4 vs action 2

Post by Jose Lozano »

Hello,

I'm running CosmoMC with the some recent cosmological measurements (CMB, BAO, PantheonPlus).

When I run CosmoMC with action 4 (testing likelihoods), the values of chi^2 are different for BAO and CMB likelihoods in comparison when I run CosmoMC with action=2 (finding best fit values). Why does this discrepancy appears? The main LCDM parameters are not fixed but vary with reasonable bounds, including cal,kgal, xi, etc.

I will appreciate if someone can give me a hint or the cause of this discrepancies of chi^2 values.

Kind regards,
Jose
Antony Lewis
Posts: 1980
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CosmoMC: chi^2 values with action 4 vs action 2

Post by Antony Lewis »

different parameters?
Jose Lozano
Posts: 19
Joined: June 27 2023
Affiliation: Universidad Nacional Autónoma de México

Re: CosmoMC: chi^2 values with action 4 vs action 2

Post by Jose Lozano »

Hello Anthony,

I'm running a dark energy model where I include 2 extra parameters plus the usual parameters (omegabh2, omegach2, ns, tau, logA, calPlanck, kgal, xi, etc).

Testing the likelihoods (action = 4) and finding the best fit (action =2), the chi^2 values of the BAO and CMB likelihoods are totally different
For example:
Chi-sq (BAO) = 12.68 (action = 4)
Chi-sq (BAO) = 38 (action = 2)

I don't get your sentence "different parameters ?"

Can you explain it a little bit more?

Kind regards,

Jose Lozano
Post Reply