To avoid contamination by non-astrophysical transients when performing population analyses on the population of binary black holes a large significance threshold must be used. However, this limits the number of signals we can analyze. To enable analyses that use a lower significance threshold, we demonstrate that the astrophysical and non-astrophysical transients can be joint fitted.

We introduce a hybrid sampling method that can significantly reduce the time to perform the large suite of conceptually similar tests of general relativity performed with gravitational-wave observations.

We show that inference of the distribution of black hole spin orientations are strongly model dependent.

The binary neutron star mass distribution measured with gravitational-wave observations has the potential to reveal information about the dense matter equation of state, supernova physics, the expansion rate of the universe, and tests of General Relativity. Most analyses of the growing population of observed binary neutron star mergers neglect the impact of the assumed neutron star spin distribution on the inferred mass distribution. We demonstrate that neglecting a poorly chosen spin distributions can seriously impact the inferred mass distribution and lead to false inferences about the most extreme neutron stars.

The traditional method used to perform population inference on Advanced LIGO/Virgo sources fails when applied to inferring the equation of state of nuclear matter from binary neutron star observations. Using Gaussian mixture modelling, we introduce a new framework in which to simultaneously infer the astrophysical distribution of merging neutron stars and the equation of state. Our method can be applied generically to population inference problems.

An introductory blog post written for the UChicago data science institute.

Sometimes repeated evaluation of interpolant is just too slow with `scipy`.