Thursday, April 22, 2010

Initial Likelihood Tests

Initial Tests of the "Improved" likelihood do not show improvements.

We seem to be missing a bunch of QSOs which we used to get. I think this is because the simulation changed from having 10M QSOs in the BOSS redshift range (2.0 < z < 3.9) to now having 10M QSOs in a larger redshift range (2.0 < z < 3.9) and most of them are low redshift, so we have a much lower density of QSOs in the redshift range of interest.

Below is a comparison of QSOs targeted with old/new methods.

The white points were targeted by both the new and old likelihoods
The magenta points were only targeted by the new likelihood
The cyan points were only targeted by the old likelihood (missed by new)

Below is a comparison of missed QSOs targeted with old/new methods.

The white points were missed by both the new and old likelihoods
The magenta points were only missed by the new likelihood
The cyan points were only missed by the old likelihood (targeted by new)

I'm going to re-run with the old luminosity function and redshift range, but with the extra QSOs as inputs to try to isolate where the problem is.

No comments:

Post a Comment