Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/facebookarchive/bootstrapped
Generate bootstrapped confidence intervals for A/B testing in Python.
https://github.com/facebookarchive/bootstrapped
Last synced: 3 months ago
JSON representation
Generate bootstrapped confidence intervals for A/B testing in Python.
- Host: GitHub
- URL: https://github.com/facebookarchive/bootstrapped
- Owner: facebookarchive
- License: other
- Archived: true
- Created: 2017-01-18T21:05:49.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2019-11-11T17:12:33.000Z (almost 5 years ago)
- Last Synced: 2024-07-13T15:09:00.589Z (4 months ago)
- Language: Python
- Size: 629 KB
- Stars: 634
- Watchers: 19
- Forks: 102
- Open Issues: 18
-
Metadata Files:
- Readme: README.rst
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
bootstrapped - confidence intervals made easy
=============================================
**bootstrapped** is a Python library that allows you to build confidence
intervals from data. This is useful in a variety of contexts - including
during ad-hoc a/b test analysis.
Motivating Example - A/B Test
-----------------------------
Imagine we own a website and think changing the color of a 'subscribe'
button will improve signups. One method to measure the improvement is to
conduct an A/B test where we show 50% of people the old version and 50%
of the people the new version. We can use the bootstrap to understand
how much the button color improves responses and give us the error bars
associated with the test - this will give us lower and upper bounds on
how good we should expect the change to be!
The Gist - Mean of a Sample
---------------------------
Given a sample of data - we can generate a bunch of new samples by
're-sampling' from what we have gathered. We calculate the mean for each
generated sample. We can use the means from the generated samples to
understand the variation in the larger population and can construct
error bars for the true mean.
bootstrapped - Benefits
-----------------------
- Efficient computation of confidence intervals
- Functions to handle single populations and a/b tests
- Functions to understand `statistical
power `__
- Multithreaded support to speed-up bootstrap computations
- Dense and sparse array support
Example Usage
-------------
.. code:: python
import numpy as np
import bootstrapped.bootstrap as bs
import bootstrapped.stats_functions as bs_stats
mean = 100
stdev = 10
population = np.random.normal(loc=mean, scale=stdev, size=50000)
# take 1k 'samples' from the larger population
samples = population[:1000]
print(bs.bootstrap(samples, stat_func=bs_stats.mean))
>> 100.08 (99.46, 100.69)
print(bs.bootstrap(samples, stat_func=bs_stats.std))
>> 9.49 (9.92, 10.36)
Extended Examples
^^^^^^^^^^^^^^^^^
- `Bootstrap
Intro `__
- `Bootstrap A/B
Testing `__
- More notebooks can be found in the
`examples/ `__
directory
Requirements
------------
**bootstrapped** requires numpy. The power analysis functions require
matplotlib and pandas.
Installation
------------
.. code:: bash
pip install bootstrapped
How bootstrapped works
----------------------
**bootstrapped** provides pivotal (aka empirical) based confidence
intervals based on bootstrap re-sampling with replacement. The
percentile method is also available.
For more information please see:
1. `Bootstrap confidence
intervals `__
(good intro)
2. `An introduction to Bootstrap
Methods `__
3. `The Bootstrap, Advanced Data
Analysis `__
4. `When the bootstrap dosen't
work `__
5. (book) `An Introduction to the
Bootstrap `__
6. (book) `Bootstrap Methods and their
Application `__
See the CONTRIBUTING file for how to help out.
Contributors
^^^^^^^^^^^^
Spencer Beecher, Don van der Drift, David Martin, Lindsay Vass, Sergey
Goder, Benedict Lim, and Matt Langner.
Special thanks to Eytan Bakshy.
License
-------
**bootstrapped** is BSD-licensed. We also provide an additional patent
grant.