Please contact sjhan@berkeley.edu or asuliga@berkeley.edu for zoom links.

Faster likelihood evaluation enhances the efficiency of gravitational wave signal analysis. We present Mode-by-mode Relative Binning (MRB), a new method designed for obtaining fast and accurate likelihoods for advanced waveform models that include spin-orbit precession effects and multiple radiation harmonics from compact binary coalescence. Leveraging the “twisting-up” procedure of constructing precessing waveform modes from non-precessing ones, the new method mitigates degrade of relative binning accuracy due to interference from superimposed modes. Additionally, we supplement algorithms for optimizing the choice of frequency bins specific to any given strain signal under analysis. Using the new method, we are able to evaluate the likelihood with up to an order of magnitude reduction in the number of waveform model calls per frequency compared to the previously used relative binning scheme, and achieve better likelihood accuracy than is sufficient for obtaining source parameter posterior distributions that are indistinguishable from the exact ones.

Please contact sjhan@berkeley.edu or anna.suliga@nbi.ku.dk for zoom links.