## 2MASS Project sanity check

I have calculated the reduced chi^2 of 341 galaxies using models from GALPHAT posterior median. This is for checking if there is any weird situation where GALPHAT model is inappropriate.

I plotted the histogram of reduced chi^2 and confirmed that it peaked around 1.2 and had a tail toward larger value. 13 out of 341 galaxies(which is 3.8%) has reduced chi^2 greater than 3.0. Therefore I checked every residual images with the reduced chi^2 > 3.0. There are several reasons for this large chi^2 per degree of freedom.

1. strong bar

2. spiral arm

3. non-axis sysmetric tail

4. ring

5. too small Sersic index. I need larger range of prior for Sersic n (0.5<n<8.0 wasn’t large enough. Maybe n>~10?)

6. need more MCMC

Advertisements