Hi guys
I'm trying to modify the PPF patch, but when I replaced w_0 and w_a with my model parameters, at the end, I got only 1-sigma confidence levels on all my parameters, and the 2-sigma is exactly as the 1-sigma.
Does anybody know what is the problem? Is it related to the defined domain of the new parameters or the way of defining them?
Thanks
Cosmomc new Parameter Problem
-
Jason Dossett
- Posts: 97
- Joined: March 19 2010
- Affiliation: The University of Texas at Dallas
- Contact:
Cosmomc new Parameter Problem
Hi,
It could actually be either one. If you are not calling the parameters correctly into your model then changing them won't do anything to your likelihood and your bounds will be basically the limits you have defined for them in params.ini
It is also equally as likely that the limits you have defined in params.ini are simply too small and thus every point within that parameter space is within the true 1 sigma interval.
The best thing to do is significantly increase the gap in your limits in params.ini. If after that the confidence levels are the same, then you might have an issue in how you are using w_0 and w_a as new model parameters.
It could actually be either one. If you are not calling the parameters correctly into your model then changing them won't do anything to your likelihood and your bounds will be basically the limits you have defined for them in params.ini
It is also equally as likely that the limits you have defined in params.ini are simply too small and thus every point within that parameter space is within the true 1 sigma interval.
The best thing to do is significantly increase the gap in your limits in params.ini. If after that the confidence levels are the same, then you might have an issue in how you are using w_0 and w_a as new model parameters.