Robust and fast parameter estimation for gravitational waves from binary neutron star merger remnants
We present a robust and efficient methodology for parameter estimation of gravitational waves generated during the post-merger phase of binary neutron star mergers. Our approach leverages an analytic waveform model combined with empirical relations to predict prior ranges for the post-merger frequencies based on measurements of the chirp mass and effective tidal deformability in the inspiral phase. This enables robust inference of the main features of the post-merger spectrum, avoiding possible multi-modality induced by wide priors. Using waveforms derived from numerical relativity, we systematically validate our model across a broad spectrum of neutron star equations of state and mass configurations, demonstrating high fitting factors. Our method can be applied in future detections of gravitational waves from the post-merger phase with third-generation gravitational wave observatories. Furthermore, by integrating the Preconditioned Monte Carlo sampling method within the pocoMC framework, we achieve substantial computational acceleration compared to conventional Bayesian techniques.