Jan 22, 2019 - Using wrong map: Skewness and Kurtosis, top hat filter with blunt edges

These results were derived using a total CIB map that included the problematic redshift slice 0-100 Mpc. As we see in the next post, this slice has been driving most of the observed non-Gaussianity.

During one of our meetings there was a worry that using top-hat filters would be problematic due to the ringing effects. To check on that, we repeated our analysis from the previous post (Jan 21st) with top hat filters that have smooth edges.

To be specific, within [math]\displaystyle{ [-\Delta, \Delta] }[/math] from the edge the filter was

[math]\displaystyle{ \frac{1}{2}\left(1 + \sin \frac{\pi (x - x_\mathrm{edge})}{2 \Delta}\right) . }[/math]

With such filter, the skewness / kurtosis results are qualitatively unchanged.

Just as an illustrative example, for comparison we repeat here the histograms from yesterday

and compare them with results obtained with the smoothed filters