Welcome to poissonregime’s documentation!

poissonregime

Poisson error bars for low count statistics, detection significances.

https://img.shields.io/pypi/v/poissonregime.svg https://api.travis-ci.com/JohannesBuchner/poissonregime.svg?branch=master&status=started Documentation Status

About

Low count statistics is not that hard. It’s just not Gaussian.

This package can answer the following questions:

  • Given k detected objects, what are the uncertainties on the true number of objects? (poissonregime.uncertainties_rate)

  • Given k “hits” out of a sample of n tries, what is the fraction and its uncertainty? (poissonregime.uncertainties_fraction)

  • What is the significance of k detections, given that I expect B background events? (poissonregime.significance)

    • what if I measure the background events from counts in a “off” region?

    • what if I have additional systematic uncertainty?

  • What is the probability distribution on the event rate, given a measured background rate? (poissonregime.posterior)

You can help by testing poissonregime and reporting issues. Code contributions are welcome. See the Contributing page.

Usage

Read the full documentation at:

https://johannesbuchner.github.io/poissonregime/

Licence

MIT.

Indices and tables