Measurement of Hubble constant: Non-Gaussian Errors in HST key project data

TL;DR

This paper uses a Kolmogorov-Smirnov test to reveal that the uncertainties in Hubble constant measurements from HST data are non-Gaussian, challenging common assumptions about error distributions.

Contribution

It introduces a novel application of the Kolmogorov-Smirnov test to assess error distributions in cosmological measurements, specifically the Hubble constant.

Findings

Uncertainties are non-Gaussian in HST measurements

Challenges the assumption of Gaussian errors in cosmology

Provides a new statistical approach for error analysis

Abstract

Assuming the Central Limit Theorem, experimental uncertainties in any data set are expected to follow the Gaussian distribution with zero mean. We propose an elegant method based on Kolmogorov-Smirnov statistic to test the above; and apply it on the measurement of Hubble constant which determines the expansion rate of the Universe. The measurements were made using Hubble Space Telescope. Our analysis shows that the uncertainties in the above measurement are non-Gaussian.