Repository with data, analysis, and plots for arXiv:2208.12872, "Accelerating Tests of General Relativity with Gravitational-Wave Signals using Hybrid Sampling" (Wolfe, Talbot, & Golomb 2022).
In this work, we use bilby>=1.0.6 and a [custom source model](custom source model) for beyond-GR waveforms to analyze simulated signals and GW150914.
Hybrid sampling is available as a ready-to-use pipeline on top of bilby_pipe; pipeline code to launch both steps of hybrid sampling (sampling first in dynesty, and then ptemcee), as well the postprocessing executable to launch ptemcee on a dynesty result file can be found here.
GW150914_results/: Posteriors for each step of the hybrid analysis of GW150914, with and without a time-maximized overlap cut; see README.md therein.
bgr_results/: Posteriors for each step of the hybrid analysis of an injected beyond-GR signal; see README.md therein.
toy-model_misspecified.ipynb, toy-model_well-specified.ipynb: Hybrid sampling analysis of the generalized Gaussian distribution when the model in the first step is incorrectly specified or correctly specified, respectively.
gw150914.ipynb, high-snr-injection.ipynb: Plots generated based on the posteriors and sampling history of our hybrid analysis of GW150914 and an injected beyond-GR signal, respectively.
config_gw150914.ini, config_injection.ini: Configuration files for bilby_pipe, with additional arguments read by our hybrid sampling pipeline.
modified_GW150914_4s.prior, modified.prior: Prior files for our analysis fo GW150914 and an injected beyond-GR signal, respectively.
toy.py: Utility statistical and plotting functions for the hybrid analysis of the generalized Gaussian distribution.
util.py: Utility functions for the analysis of GW150914 and an injected signal, including plotting functions and common formatting definitions.
tgr-hybrid-sampling.mplstyle: Matplotlib style file for plot creation.