VAMANA: Modeling Binary Black Hole Population with Minimal Assumptions. (arXiv:2006.15047v3 [astro-ph.HE] UPDATED)
<a href="http://arxiv.org/find/astro-ph/1/au:+Tiwari_V/0/1/0/all/0/1">Vaibhav Tiwari</a>
The population analysis of compact binaries involves the reconstruction of
some of the gravitational wave (GW) signal parameters, such as, the mass and
the spin distribution, that gave rise to the observed data. This article
introduces VAMANA, which reconstructs the binary black hole population using a
mixture model and facilitates excellent density measurement as informed by the
data. VAMANA uses a mixture of weighted Gaussians to reconstruct the chirp mass
distribution. We expect Gaussian mixtures to provide flexibility in modeling
complex distributions and enable us in capturing details in the astrophysical
chirp mass distribution. Each of the Gaussian in the mixture is combined with
another Gaussian and a power-law to simultaneously model the spin component
aligned with the orbital angular momentum and the mass ratio distribution, thus
also allowing us to capture their variation with the chirp mass. Additionally,
we can also introduce broadband smoothing by restricting the Gaussian mixture
to lie within a threshold distance of a predefined reference chirp mass
distribution. Using simulated data we show the robustness of our method in
reconstructing complex populations for a large number of observations. We also
apply our method to the publicly available catalog of GW observations made
during LIGO’s and Virgo’s first and second observation runs and present the
reconstructed mass, spin distribution, and the estimated merger rate of binary
black holes.
The population analysis of compact binaries involves the reconstruction of
some of the gravitational wave (GW) signal parameters, such as, the mass and
the spin distribution, that gave rise to the observed data. This article
introduces VAMANA, which reconstructs the binary black hole population using a
mixture model and facilitates excellent density measurement as informed by the
data. VAMANA uses a mixture of weighted Gaussians to reconstruct the chirp mass
distribution. We expect Gaussian mixtures to provide flexibility in modeling
complex distributions and enable us in capturing details in the astrophysical
chirp mass distribution. Each of the Gaussian in the mixture is combined with
another Gaussian and a power-law to simultaneously model the spin component
aligned with the orbital angular momentum and the mass ratio distribution, thus
also allowing us to capture their variation with the chirp mass. Additionally,
we can also introduce broadband smoothing by restricting the Gaussian mixture
to lie within a threshold distance of a predefined reference chirp mass
distribution. Using simulated data we show the robustness of our method in
reconstructing complex populations for a large number of observations. We also
apply our method to the publicly available catalog of GW observations made
during LIGO’s and Virgo’s first and second observation runs and present the
reconstructed mass, spin distribution, and the estimated merger rate of binary
black holes.
http://arxiv.org/icons/sfx.gif
