https://github.com/jamalsenouci/causalimpact

Python port of CausalImpact R library
https://github.com/jamalsenouci/causalimpact

Last synced: 13 days ago
JSON representation

Python port of CausalImpact R library

• Host: GitHub
• URL: https://github.com/jamalsenouci/causalimpact
• Owner: jamalsenouci
• License: apache-2.0
• Created: 2016-04-04T05:57:46.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2024-04-22T00:56:40.000Z (7 months ago)
• Last Synced: 2024-09-28T09:18:20.784Z (about 1 month ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 5.4 MB
• Stars: 268
• Watchers: 20
• Forks: 63
• Open Issues: 8
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE.md
  • Authors: AUTHORS.md

Awesome Lists containing this project

README

        
## CausalImpact

[![Python package](https://github.com/jamalsenouci/causalimpact/actions/workflows/main.yml/badge.svg)](https://github.com/jamalsenouci/causalimpact/actions/workflows/main.yml)

[![codecov](https://codecov.io/gh/jamalsenouci/causalimpact/branch/master/graph/badge.svg?token=EIPC36VQHS)](https://codecov.io/gh/jamalsenouci/causalimpact)

![monthly downloads](https://pepy.tech/badge/causalimpact/month)

[![DeepSource](https://deepsource.io/gh/jamalsenouci/causalimpact.svg/?label=active+issues&show_trend=true&token=R5aIDSkIId_5THWTAPKccjcH)](https://deepsource.io/gh/jamalsenouci/causalimpact/?ref=repository-badge)

#### A Python package for causal inference using Bayesian structural time-series models

This is a port of the R package CausalImpact, see: https://github.com/google/CausalImpact.

This package implements an approach to estimating the causal effect of a designed intervention on a time series. For example, how many additional daily clicks were generated by an advertising campaign? Answering a question like this can be difficult when a randomized experiment is not available. The package aims to address this difficulty using a structural Bayesian time-series model to estimate how the response metric might have evolved after the intervention if the intervention had not occurred.

As with all approaches to causal inference on non-experimental data, valid conclusions require strong assumptions. The CausalImpact package, in particular, assumes that the outcome time series can be explained in terms of a set of control time series that were themselves not affected by the intervention. Furthermore, the relation between treated series and control series is assumed to be stable during the post-intervention period. Understanding and checking these assumptions for any given application is critical for obtaining valid conclusions.

#### Try it out in the browser

[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/jamalsenouci/causalimpact/HEAD?labpath=GettingStarted.ipynb)

#### Installation

install the latest release via pip

```bash

pip install causalimpact

```

#### Getting started

[Documentation and examples](https://nbviewer.org/github/jamalsenouci/causalimpact/blob/master/GettingStarted.ipynb)

#### Further resources

- Manuscript: [Brodersen et al., Annals of Applied Statistics (2015)](http://research.google.com/pubs/pub41854.html)

#### Bugs

The issue tracker is at https://github.com/jamalsenouci/causalimpact/issues. Please report any bugs that you find. Or, even better, fork the repository on GitHub and create a pull request.