Skip to content

spotify/confidence

Repository files navigation

Spotify Confidence

Status Latest release Python Python Python Python

Python library for AB test analysis.

Why use Spotify Confidence?

Spotify Confidence provides convenience wrappers around statsmodel's various functions for computing p-values and confidence intervalls. With Spotify Confidence it's easy to compute several p-values and confidence bounds in one go, e.g. one for each country or for each date. Each function comes in two versions:

  • one that return a pandas dataframe,
  • one that returns a Chartify chart.

Spotify Confidence has support calculating p-values and confidence intervals using Z-statistics, Student's T-statistics (or more exactly Welch's T-test), as well as Chi-squared statistics. It also supports a variance reduction technique based on using pre-exposure data to fit a linear model.

There is also a Bayesian alternative in the BetaBinomial class.

Basic Example

import spotify_confidence as confidence
import pandas as pd

data = pd.DataFrame(
    {'variation_name': ['treatment1', 'control', 'treatment2', 'treatment3'],
     'success': [50, 40, 10, 20],
     'total': [100, 100, 50, 60]
    }
)

test = confidence.ZTest(
    data,
    numerator_column='success',
    numerator_sum_squares_column=None,
    denominator_column='total',
    categorical_group_columns='variation_name',
    correction_method='bonferroni')
    
test.summary()
test.difference(level_1='control', level_2='treatment1')
test.multiple_difference(level='control', level_as_reference=True)

test.summary_plot().show()
test.difference_plot(level_1='control', level_2='treatment1').show()
test.multiple_difference_plot(level='control', level_as_reference=True).show()

There's a lot more you can do:

  • Segment results by one or more dimensions
  • Use non-inferiority margins
  • Group sequential tests
  • Sample size and power calculations
  • etc

See jupyter notebooks in examples folder for more complete examples.

Installation

Spotify Confidence can be installed via pip:

pip install spotify-confidence

Find the latest release version here

Code of Conduct

This project adheres to the Open Code of Conduct By participating, you are expected to honor this code.