Computation of variable Theta, not self-consistent ?

[Karsten Jedamzik]

Hi,

It seems to me that the variable Theta is computed by using the Hu & Sugiyama
recombination redshift zstar approximation, as opposed to computing zstar directly
from CAMB (recfast). Is this correct ?

If correct, that implies that when derived parameters, for example H, are computed
the error in the zstar approximation propagates into H. This would be particularly
bad for alternative models where zstar does not follow the Hu & Sugiyama approximation.

I assume however, that the liklihood of a particular parameter combination is computed
directly from the observational data, and the CAMB results, such that a posteriori
probabilities are correctly computed, independent of the zstar assumption.

Could somebody comment, thanks

Karsten

[jayanti prasad]

I also want to ask the same.. check my last question...