Periodicity analysis for Michilli et al, "Sub-second periodicity in a fast radio burst", https://arxiv.org/abs/2107.08463 Note: these scripts take 12+ hours to run and generate ~80 GB of data! The scripts perform auxiliary analyses which are not in the paper, and they also use 10^8 Monte Carlo simulations per analysis. (It would be easy to modify the scripts to run faster by doing fewer MCs and auxiliary analysis; let me know if that would be useful.) ---------------------------------------------------------------------------------------------------- ./01-make-initial-plots.py Makes some quick exploratory plots from the CHIME ToAs ("vertical" and "regression" plots). The plots which are relevant for the paper are: plots_v2/event1_p9_vertical.pdf plots_v2/event1_p9_regression.pdf (*) plots_v2/event2_vertical.pdf plots_v2/event2_regression.pdf plots_v2/event3_vertical.pdf plots_v2/event3_regression.pdf (*) Appears in paper as "Extended Data Figure 3" ---------------------------------------------------------------------------------------------------- ./02-make-sims.py A long-running script (12+ hours) which generates a ton (~80 GB) of simulated ToAs. (The CHIME ToA's are not used.) The output files have filenames such as: sims_n100000000/sim_p9_g5_chi0.2.npy (In code, you can call toa_periodicity.sim_filename()) ---------------------------------------------------------------------------------------------------- ./03-postprocessing.py Evaluates data S-hat statistics, compares with sims, computes p-values. Results written here: analysis_v2/pvalues_n100000000.pkl Writes some S-hat histograms, with filenames such as plots_v2/event1_p9_maxgap5_histogram.pdf ---------------------------------------------------------------------------------------------------- ./04-plots.py A summary table of results is written here: plots_v2/summary.txt (*) (*) Selected results from this table appear in "Extended Data Table 2" in the paper. ---------------------------------------------------------------------------------------------------- ./05-daniele-plot.py Creates the following plots: plots_v2/histogram_event1_p9.pdf plots_v2/histogram_event2.pdf plots_v2/histogram_event3.pdf These plots are shown in the right column of Figure 2 in the paper.