Statistical software · statsmodels developers (NumFOCUS-sponsored open-source project)

statsmodels

Open-source Python library for classical statistical modeling, hypothesis testing, and econometrics alongside pandas and scikit-learn.

Visit statsmodels Alternatives

statsmodels is a free, open-source Python library focused on classical, inference-oriented statistics rather than predictive machine learning. It provides ordinary and generalized linear models, robust regression, time-series models (ARIMA, SARIMAX, state-space methods), and a wide range of hypothesis tests and diagnostic checks, each with the coefficients, standard errors, and p-values that applied statisticians and econometricians need. Its R-like formula API lets analysts specify models the way they would in R, and it integrates directly with pandas DataFrames. Where scikit-learn is optimized for prediction accuracy, statsmodels is optimized for statistical inference — understanding why a relationship exists and how confident to be in it — making it the standard choice for econometrics and hypothesis-driven analysis within the Python data-science stack.

At a glance

Vendor statsmodels developers (NumFOCUS-sponsored open-source project)
Pricing model Open source + paid options
Free tier Yes
Deployment Self-hosted
Open source Yes (BSD-3-Clause)
Best for Python-based analysts and econometricians who need rigorous statistical inference, not just predictive accuracy.

Pricing

Free and open source; no commercial tiers.

Pricing has not been verified yet — see the vendor's site.

Features

  • OLS, GLM, and robust regression
  • Time-series models (ARIMA, SARIMAX, state-space)
  • Hypothesis tests (t-tests, ANOVA, nonparametric tests)
  • Model diagnostics and specification tests
  • Generalized estimating equations
  • R-like formula API
  • Direct integration with pandas DataFrames

Integrations

Profile last reviewed September 21, 2026

Head to head

Alternatives

SciPy Open-source Python library of numerical algorithms — optimization, integration, statistics, linear algebra — underlying the scientific Python stack. R Open-source programming language and environment for statistical computing and graphics, the default language of applied statistics education. Julia Open-source, high-performance programming language for numerical and scientific computing, used for statistical simulation at near-C speed.

statsmodels in the index now

Terms to know

Regression analysis A statistical method for estimating the relationship between one outcome variable and one or more predictor variables. P-value The probability of observing data at least as extreme as the result, if the null hypothesis were actually true. Statistical significance A threshold judgment that an observed effect is unlikely to be due to chance alone, given the data and test used. Time series forecasting Predicting future values of a metric using patterns found in its own historical sequence over time. Hypothesis testing A statistical framework for deciding whether observed data provides enough evidence to reject a default assumption. ARIMA (autoregressive integrated moving average) A classical time-series forecasting model that combines a series' own past values, differencing, and past forecast errors.

Related guides

How to choose statistical software The decisions that matter are code vs menus, general-purpose vs domain specialist, and whether your field already has a de facto standard you can't ignore.