Code to compute the individual SN covariance matrices in JLA (Betoule+ 2014 Section 5.5)
To run the scripts you need a configuration file and list of SNe in the Covariance/scripts/ directory. Here are a few lines from each of mine (for the JLA sample) to indicate the format:
JLA.config:
lightCurves $JLA/data/jla_light_curves/ # the JLA lightcurves lightCurveFits $JLA/data/JLA_20160531.fits # fits file containing results of lightcurve fits
where I have a $JLA environmental variable pointing to where these data are. For individual scripts you may need other keys in this configuration file - contact Bonnie or Chris if you are trying to run a script and something is missing.
JLA.list:
lc-03D1au.list LC $JLA/data/jla_light_curves_dm/new_lc-03D1au.list lc-03D1aw.list LC $JLA/data/jla_light_curves_dm/new_lc-03D1aw.list lc-03D1ax.list LC $JLA/data/jla_light_curves_dm/new_lc-03D1ax.list
Scripts:
- jla_compute_C*.py
code for C_* for each contribution e.g. host, dust, bias, nonia, model, cal, pecvel, stat
general usage is 'python jla_compute_C*.py -s JLA.list' (or other list of SNe). Some of these require other arguments and input. We're working on making these available; in the meantime email us. See docstring in each code for technical details of each calculation.
- jla_compute_diag_terms.py
computes diagonal terms in magnitude covariance matrix (note: all other covariance matrices are in eta i.e. m_B, X1, C) due to lensing, scatter in peculiar velocity and intrinsic scatter
- jla_compute_rel_size.py
computes relative contributions of various covariance matrices (each systematic and statistics) according to method in B14 Section 6.2
- JLA_library.py
library of common functions used in above scripts
The following scripts aren't essential parts of the chain:
- jla_merge_lightcurve_fits.py
formats lightcurves from multiple files and/or samples into .fits files
- jla_compute_DateofMax.py
inserts date of maximum into light curve files
- jla_compute_ZP.py
computes magnitude of CALSPEC standards