Type Ia Supernova Intrinsic Magnitude Dispersion and the Fitting of Cosmological Parameters

TL;DR

This paper introduces a new method for fitting cosmological parameters using Type Ia supernova data that simultaneously estimates intrinsic magnitude dispersion within a single, unbiased statistical framework, improving accuracy and reliability.

Contribution

It presents a unified, non-iterative analysis method that estimates intrinsic dispersion and cosmological parameters together, enhancing statistical robustness and bias reduction.

Findings

The method provides unbiased estimates of intrinsic dispersion.

Parameter uncertainties include dispersion uncertainty.

Differences with existing methods are negligible for current data.

Abstract

I present an analysis for fitting cosmological parameters from a Hubble Diagram of a standard candle with unknown intrinsic magnitude dispersion. The dispersion is determined from the data themselves, simultaneously with the cosmological parameters. This contrasts with the strategies used to date. The advantages of the presented analysis are that it is done in a single fit (it is not iterative), it provides a statistically founded and unbiased estimate of the intrinsic dispersion, and its cosmological-parameter uncertainties account for the intrinsic dispersion uncertainty. Applied to Type Ia supernovae, my strategy provides a statistical measure to test for sub-types and assess the significance of any magnitude corrections applied to the calibrated candle. Parameter bias and differences between likelihood distributions produced by the presented and currently-used fitters are negligibly small for existing and projected supernova data sets.