Hierarchical Inference of Binary Neutron Star Mass Distribution and Equation of State with Gravitational Waves

TL;DR

This paper introduces a new statistical framework using Gaussian mixture models to jointly infer the mass distribution of binary neutron stars and their nuclear equation of state from gravitational wave data, improving accuracy and robustness.

Contribution

The authors develop a novel method that simultaneously estimates neutron star mass distribution and the equation of state, addressing limitations of previous approaches.

Findings

Reproduces expected precision with tens of observations

Mismodeling the equation of state biases mass distribution inference

Method is broadly applicable to population inference problems

Abstract

Gravitational wave observations of binary neutron star mergers provide valuable information about neutron star structure and the equation of state of dense nuclear matter. Numerous methods have been proposed to analyze the population of observed neutron stars and previous work has demonstrated the necessity of jointly fitting the astrophysical distribution and the equation of state in order to accurately constrain the equation of state. In this work, we introduce a new framework to simultaneously infer the distribution of binary neutron star masses and the nuclear equation of state using Gaussian mixture model density estimates which mitigates some of the limitations previously-used methods suffer from. Using our method, we reproduce previous projections for the expected precision of our joint mass distribution and equation of state inference with tens of observations. We also show that mismodeling the equation of state can bias our inference of the neutron star mass distribution. While we focus on neutron star masses and matter effects, our method is widely applicable to population inference problems.