This package implements a hierarchical bayesian framework for constraining the Hubble parameter and modified GW propagation with dark sirens and galaxy catalogues.
The methods and results can be found in the paper Cosmology with LIGO/Virgo dark sirens: Hubble parameter and modified gravitational wave propagation.
Developed by Andreas Finke and Michele Mancarella.
This package is released together with the paper Cosmology with LIGO/Virgo dark sirens: Hubble parameter and modified gravitational wave propagation. When making use of it, please cite the paper and the present git repository. Bibtex:
@misc{finke2021cosmology,
title={Cosmology with LIGO/Virgo dark sirens: Hubble parameter and modified gravitational wave propagation},
author={Andreas Finke and Stefano Foffa and Francesco Iacovelli and Michele Maggiore and Michele Mancarella},
year={2021},
eprint={2101.12660},
archivePrefix={arXiv},
primaryClass={astro-ph.CO}
}
First, in a terminal run
pip install -r requirements.txt
to install the required python libraries. Then, run
./install.sh
the code will download all the needed data in the data directory (for its structure, go to data/). These include:
- The GLADE galaxy catalogue
- O2 and O3 skymaps from the LVC official data releases
- O2 and O3 strain sensitivities
- Optionally, the DES and GWENS galaxy catalogues. Do not use this option on a laptop, since the space required is very large
Here we will soon provide a description of the main logic of the code
Description is provided inside the data folder.
The configuration options are read from the the file config.py .
We provide a template with explanation in config_template.py. To creat you own configuration file:
cp config_template.py config.py
Then, open config.py and set the options. A description is provided within the file.
The default options are for running inference for H0 on the O3 BBH events, with flat prior between 20 and 140, mask completeness with 9 masks, interpolation between multiplicative and homogeneous completion, B-band luminosity weights, and a completeness threshold of 50%. The selection effects are computed with MC. To run, execute
python main.py
The result will be saved in a folder results/O2BBHs/ (the name can be changed in the configuration).