# alkaline-ml/pmdarima

A statistical library designed to fill the void in Python's time series analysis capabilities, including the equivalent of R's auto.arima function.

Repository: https://github.com/alkaline-ml/pmdarima
Canonical: https://ross.abutalabs.com/products/pmdarima
Homepage: https://www.alkaline-ml.com/pmdarima
Language: Python
License: MIT
License Family: permissive
Topics: arima, time-series, forecasting, forecasting-models, python, econometrics, pmdarima, machine-learning, sarimax
Last push: 2025-11-17T16:09:28+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 52, release rhythm 57, longevity 100
- inputs: {"age_days": 3443, "days_push": 289, "days_rel": 289, "gap_med": 4, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1733, forks 252 (observed 2026-08-28T04:05:28.956140+00:00)

## What it is
pmdarima is a Python statistical library that brings R's auto.arima functionality to Python, wrapping statsmodels ARIMA/SARIMAX models behind a scikit-learn-style interface. It also provides stationarity and seasonality tests, time series utilities, transformations, cross-validation, and built-in datasets.

## Use cases
- automatically fit ARIMA models to time series in Python
- replace R's forecast::auto.arima with a Python equivalent
- forecast seasonal time series with SARIMAX
- run stationarity tests like augmented Dickey-Fuller on a series
- difference and inverse-difference time series data
- build scikit-learn pipelines for time series forecasting
- cross-validate time series forecasting models
- decompose seasonal time series

## When to choose
- you need automatic ARIMA order selection (auto.arima) in Python
- you want a scikit-learn-like API for classical time series forecasting
- you need statistical tests for stationarity and seasonality
- you're migrating forecasting workflows from R to Python

## When to avoid
- you need deep learning or modern neural forecasting models
- you prefer Prophet or gradient-boosting-based forecasting
- your project requires very recent Python versions beyond the library's supported range
- you need actively developed features — check maintenance status, as development has slowed

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science, math, sdk
- domain: data-science, time-series, machine-learning, analytics
- platform: python, cross-platform, windows
- tags: arima, auto-arima, forecasting, sarimax, statsmodels, scikit-learn, time-series-analysis, stationarity-tests, seasonality, econometrics, linux, macos

## Member repositories
- alkaline-ml/pmdarima (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:28.956140+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:31:26.961736+00:00, confidence not recorded.
  - readme: https://github.com/alkaline-ml/pmdarima (fetched 2026-08-28T04:05:28.956140+00:00, sha 6e91e942df06)
  - homepage: https://www.alkaline-ml.com/pmdarima (fetched 2026-08-29T11:08:14.436896+00:00, sha 7ebf06f69680)
  - site_page: https://alkaline-ml.com/pmdarima/about.html (fetched 2026-08-29T11:08:14.445846+00:00, sha 390eb7a19856)
  - site_page: https://alkaline-ml.com/pmdarima/quickstart.html (fetched 2026-08-29T11:08:14.447756+00:00, sha 1371349c99a6)
  - registry_pypi: https://pypi.org/pypi/pmdarima/json (fetched 2026-08-29T11:08:14.449460+00:00, sha cf15ec1dee8c)
- Data as of 2026-08-30T08:39:29.467469+00:00.