XENON1T / laidbax

Likelihood And Interpolated-Density Based Analysis for XENON

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Likelihood- And Interpolated Density Based Analysis for Xenon

Jelle Aalbers, 2016

Statistical XENON1T Analysis for the lazy analyst.

Source code: https://github.com/XENON1T/laidbax

Documentation:

About

Did you spend years of your life in a dark cave below the mountain building that awesome TPC? Do you want to spend the next few years behind a dimly lit computer screen making sense of the data that comes out? Then this package is not for you.

This package allows you to do parametric inference using Monte-Carlo derived extended unbinned likelihood functions. It lets you make likelihood functions which measure agreement between data and Monte Carlos with different settings: you choose which settings to vary (which parameters the likelihood functions has) and in which space the agreement is measured. For more information, please see the documentation in this note and the examples in the notebooks folder.

All of the hard work is done by blueice; this only contains the code necessary to make blueice work with a simple Monte Carlo of the XENON1T experiment. Much of this is derived or shamelessly "borrowed" from other sources:

  • Andrew's maximum-gap limit setting code, used for the XENON100 max-gap cross checks.
  • Chris' wimpstat repository, used for the XENON100 S2-only limit setting (but no longer available?)
  • NEST: not directly, but since this is currently the best xenon TPC code out there, the physics model used here
  • Several of the Monte Carlo group's excellent material on this topic, in particular the XENON1T Monte Carlo paper and the notes linked here.

The default model includes lots of information I obtained by curve-tracing plots or even just constructing some function that looked like what I saw in an image. Just saying...

NB: These are NOT the official XENON1T models used for the first results paper (https://arxiv.org/abs/1705.06655).

Installation

First install the major dependencies: blueice and pax. We need pax for the common unit system and XENON1T.ini.

Then run python setup.py develop or python setup.py install

About

Likelihood And Interpolated-Density Based Analysis for XENON

License:Other


Languages

Language:Jupyter Notebook 98.2%Language:Python 1.8%