Supplementary material to "Constraints on Binary Black Hole Populations from LIGO-Virgo Detections"

This repository contains the code used to reanalyze the LIGO-Virgo data and do the population inference described in arxiv.org/abs/1806.10610

It is free to use.

Purpose

1. Estimate parameter likelihood of BBH merger events using a grid in parameter space (chirp mass, mass ratio, effective spin).

2. Estimate horizon distance dependence on parameters to account for selection effects.

3. Infer population parameters describing the BBH merger rate dependence on the BBH parameters.

Requisites

pycbc, jupyter

Usage

1-estimate_parameters

• Write a file <event>.dat whose columns are the detector strains sampled at 4096 Hz, and header H1 L1 ...
• Write a file <event>_ligo_parameters with the parameters LIGO reported for that event.
  Tip: Edit and run source2detector_frame.py if LIGO only reported source-frame quantities.
• Run python config_parameters.py <event> for the desired event (generates grid in parameter space). Alternatively, run bash all_config to do all events at once.
• Run python compute_likelihood.py <event> within the pycbc environment (generates templates on the grid and matches them to the data). Alternatively, run bash all_compute to do all events at once.
• Optional: analyze the event data with estimate_parameters.ipynb

2-estimate_horizon

• Run config_parameters.py (generates grid in parameter space).
• Run compute_SNR_1Mpc.py all_par_space/ with pycbc (generates templates on the grid and stores the expected SNR of an optimally-aligned source at 1 Mpc, given a reference amplitude spectral density (ASD)).

3-analyze_population

• Run analyze_population.ipynb

File structure

.
├── 1-estimate_parameters/              # Single-event analysis
│   ├── all_clear.sh                    # Clear all the runs (run 1st)
│   ├── all_compute.sh                  # Run compute_likelihood.py on all the events (run 3rd)
│   ├── all_config.sh                   # Run parameter_config.py on all the events (run 2nd)
│   ├── compute_likelihood.log2
│   ├── compute_likelihood.py           # Compute parameter likelihood over parameter_grid
│   ├── estimate_parameters.ipynb       # Analyze the single-event likelihoods
│   ├── logI.dat                        # Tabulated function log(I(|z|))
│   ├── logI.nb                         # Generate logI.dat
│   ├── parameter_config.py             # Generate parameter_grid and grid_metadata
│   ├── figures/                        # Multi-event figures
│   ├── <event1>/
│   │   ├── figures/                    # Single-event figures
│   │   ├── freqs.dat                   # Template frequencies (Hz)
│   │   ├── grid_metadata               # Dimensions of the parameter grid
│   │   ├── <event1>.dat                # Strain data
│   │   ├── <event1>_LIGO_parameters    # Parameter values reported by the LVC
│   │   └── parameter_grid              # Parameter values on which to estimate the likelihood
│   └── .../
│
├── 2-estimate_horizon/
│   ├── ASD_average.dat                 # A reference amplitude spectral density
│   ├── ASD_zdhp.dat
│   ├── compute_PSD_average.py          # Generate ASD_average.dat
│   ├── compute_SNR_1Mpc.py             # Compute the SNR over parameter_grid
│   └── all_par_space/
│       ├── figures/
│       ├── freqs.dat
│       ├── grid_metadata
│       ├── parameter_config.py
│       └── parameter_grid
│
├── 3-analyze_population
│   ├── analyze_population.ipynb        # Do population inference
│   ├── P_greater_than_w.dat            # Tabulated cumulative of angular factors, P(w > w*)
│   ├── P_greater_than_w.ipynb          # Generate P_greater_than_w.dat
│   └── figures/
│
└── README.md                           # This file

---

Downloading data

Download the event data in txt format from https://losc.ligo.org/events and merge the detector files, e.g. with

`paste H1_file L1_file ... > event.dat`

Then edit the headers so they read H1 L1 ...

Pycbc

Install e.g. with

docker pull pycbc/pycbc-el7:latest

Run e.g. with

docker run -e DISPLAY=${DISPLAY} -v /tmp/.X11-unix:/tmp/.X11-unix/ -v ${HOME}/<path/to>/constraints_bbh_populations:/home/pycbc/constraints_bbh_populations -v ${HOME}/.ssh:/home/pycbc/.ssh -it pycbc/pycbc-el7:latest /bin/bash -l