Application of the Kolmogorov-Smirnov test to CMB data: Is the universe really weakly random?

TL;DR

This paper reevaluates the application of the Kolmogorov-Smirnov test to CMB data, finding no evidence that the universe's CMB signal is weakly random, contrary to previous claims, due to proper consideration of correlations.

Contribution

It demonstrates the importance of accounting for correlations in the KS test when analyzing CMB data, challenging prior claims of non-randomness.

Findings

No deviation from the standard {1}CDM model was found.

Proper correlation consideration explains previous apparent anomalies.

Supports the view that CMB is consistent with Gaussian randomness.

Abstract

A recent application of the Kolmogorov-Smirnov test to the WMAP 7 year W-band maps claims evidence that the CMB is "weakly random", and that only 20% of the signal can be explained as a random Gaussian field. I here repeat this analysis, and in contrast to the original result find no evidence for deviation from the standard {\Lambda}CDM model. Instead, the results of the original analysis are consistent with not properly taking into account the correlations of the {\Lambda}CDM power spectrum.