SN and BAO constraints on (new) polynomial dark energy parametrizations: current results and forecasts

TL;DR

This paper introduces two new polynomial dark energy parametrizations, demonstrating their improved performance over existing models using current and forecasted BAO and supernova data, with better correlation properties and tighter constraints.

Contribution

The paper proposes novel polynomial dark energy parametrizations with low correlation and improved fit, validated through simulations and comparison with the CPL model.

Findings

Our models outperform CPL in Bayesian DIC and Figure-of-Merit.

They exhibit lower parameter correlation and smaller errors.

Simulations show better constraints with future BAO surveys.

Abstract

In this work we introduce two new polynomial parametrizations of dark energy and explore their correlation properties. The parameters to fit are the equation of state values at z=0 and z=0.5, which have naturally low correlation and have already been shown to improve the popular Chevallier-Polarski-Linder (CPL) parametrization. We test our models with low redshift astronomical probes: type Ia supernovae and baryon acoustic oscillations (BAO), in the form of both current and synthetic data. Specifically, we present simulations of measurements of the radial and transversal BAO scales similar to those expected in a BAO high precision spectroscopic redshift survey similar to EUCLID. According to the Bayesian deviance information criterion (DIC), which penalizes large errors and correlations, we show that our models perform better than the CPL re-parametrization proposed by Wang (in terms of z=0 and z=0.5). This is due to the combination of a lower correlation and smaller relative errors. The same holds for a frequentist perspective: our Figure-of-Merit is larger for our parametrizations.