Skip to main content

Basic hypothesis testing made simple

Project description

https://img.shields.io/pypi/v/medstat.svg https://img.shields.io/travis/monferrand/medstat.svg License Documentation Status

medstat is a library aiming to make basic hypothesis testing as simple as possible.

Getting started

This project is available on PyPI you can just:

pip install medstat

Quick Example

Load your data in a dataframe using for instance pd.read_csv() or pd.read_excel().

data = pd.read_csv("my_data.csv")

Test a single hypothesis:

>>> medstat.test_hypothesis(data, 'sex', 'age < 30')

{'contingency_table':
age < 30  False  True  All
sex
Female       26    22   48
Male         24     8   32
All          50    30   80,
'test': 'Fisher',
'p-value': 0.06541995357625573,
'significant': False}

Or test many hypothesis at the same time:

result = medstat.analyse_dataset(data,
                                 [('sex', 'age < 30'),
                                  ('sex', 'test_a'),
                                  ('test_a', 'age > 50'),
                                 ])

It prints the output:

-------------------- Test 1 --------------------
Test independence between sex and age < 30.
Use Chi-squared test.
Result is not significant.
p-value: 0.18407215636751517
Contingency table:
 age < 30  False  True  All
sex
Female       21    18   39
Male         29    12   41
All          50    30   80


-------------------- Test 2 --------------------
Test independence between sex and test_a.
Use Chi-squared test.
Result is not significant.
p-value: 0.9539453144224308
Contingency table:
 test_a  negative  positive  All
sex
Female        25        14   39
Male          25        16   41
All           50        30   80


-------------------- Test 3 --------------------
Test independence between test_a and age > 50.
Use Fisher test.
Result is significant.
p-value: 6.392910983822276e-12
Contingency table:
 age > 50  False  True  All
test_a
negative     46     4   50
positive      5    25   30
All          51    29   80

You can also save it to a text file using the file argument.

result = medstat.analyse_dataset(data,
                                 [('sex', 'age < 30'),
                                  ('sex', 'test_a'),
                                  ('test_a', 'age > 50'),
                                 ],
                                file='report.txt')

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

medstat-1.0.2.tar.gz (12.4 kB view hashes)

Uploaded Source

Built Distribution

medstat-1.0.2-py2.py3-none-any.whl (5.3 kB view hashes)

Uploaded Python 2 Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page