Frequently asked questions
Short answers to the questions people ask most often about OpenEconometrics. If yours is not here, open an issue on GitHub.
What is OpenEconometrics?
OpenEconometrics is an open-source Python library for statistics and econometrics, published on PyPI as openecon, together with a desktop workbench that has a code editor, a persistent Python session, a data inspector, charts and LaTeX export. Estimators are implemented in the project itself on PyTorch float64 tensors.
Is OpenEconometrics free?
Yes. The code is free and open source under the Apache 2.0 licence, including for commercial use. Optional cloud, team and institutional support services may be offered in the future; the licensing policy describes the plan.
Can it replace Stata, EViews or SPSS?
For some projects, yes, and for others not yet. OpenEconometrics covers many standard estimators with Stata-style names, robust and clustered inference, postestimation and publication tables. Full Stata coverage and numerical parity with Stata have not been established, and it does not run Stata, EViews or SPSS commands. The workflow guide lists what is covered and the current limits.
Which methods are supported?
The source-generated capability inventory lists 171 registered estimator names, including linear and panel regression, instrumental variables and GMM, discrete-choice and count models, quantile regression, time-series models, survival analysis and treatment-effect estimators. Each estimator lists its supported options and exclusions. In Python, oe.capabilities() returns the same information.
Which platforms does it run on?
The Python library supports Python 3.11 to 3.14. The desktop alpha is available for Apple Silicon Macs with macOS 15 or later, and for Windows. See the download page.
Is the Mac app notarized?
No. The Mac alpha is ad-hoc signed. Apple Developer ID signing and notarization are pending, so macOS asks you to confirm the first launch. The download page shows the steps and a checksum you can verify. The Windows installer is unsigned.
Where does my data go?
Analyses run on your own computer, and data you import stays in your local workspace. If you use the team edition, the drafts, data and results you choose to share are stored in your team's cloud project, and the cloud service never runs code. The website loads Google Analytics only if you accept statistics cookies; see the privacy notice.
Can I use it with an AI agent?
Yes. OpenEconometrics includes an MCP server that lets Codex, Claude and other MCP clients inspect datasets, check capabilities and fit supported models through structured tools. It does not expose arbitrary code execution. Summaries and model output are shared with the client, which may forward them to its AI provider, and raw rows are shared only when explicitly requested.
How accurate are the results?
Estimates are computed in 64-bit floating point and tested against independent references, such as certified datasets, analytical results and published examples. For example, OLS agrees with the NIST Longley certified values to at least 14 significant digits. Passing such tests does not prove every method and option correct. The guide on validating econometric results explains how to check any package.
Can I use the teaching labs in my course?
Yes. The 80 teaching labs in econometrics, statistics, microeconomics and advanced econometrics are free under Apache 2.0. Student and instructor editions are available as downloads.
How do I report a bug or suggest a feature?
Open an issue in the GitHub repository. Do not put credentials, personal data or private files in a public issue. To report a security vulnerability, follow the security policy.
Try it on your own data
OpenEconometrics is free and open source under Apache 2.0. Install the Python library or download the desktop alpha.
pip install openecon