Description
statsmodels is an agent skill for fitting and diagnosing statistical models in Python with the statsmodels library. It covers regression, time series and econometric models, including OLS, GLM, discrete and mixed models, ARIMA and SARIMAX. It also covers coefficient inference, marginal effects and model diagnostics. The skill is clear about what a fit shows. A successful fit establishes that the numerical procedure ran, but causal identification, calibrated uncertainty and model adequacy depend on a defensible study design and assumptions. The skill was reviewed against statsmodels 0.15.0, released on 27 August 2026. Native checks used Python 3.13, NumPy 2.5.3, SciPy 1.18.1, pandas 3.0.6, matplotlib 3.11.2 and scikit-learn 1.9.1, in a dedicated environment. It recommends the stable high-level imports from statsmodels.api and statsmodels.formula.api, with direct module imports where newer or specialised classes are needed. In version 0.15, the skill advises using result_object=True with named fields for the ADF and KPSS tests, and rng= where older versions accepted seed or random_state.…
Highlights
- OLS, GLM, discrete and mixed models
- ARIMA and SARIMAX time series
- Coefficient inference and diagnostics
- Stable statsmodels.api imports
- Reviewed on statsmodels 0.15.0
Ratings & Reviews
No ratings yet
Be the first to rate statsmodels.
People also visit
statsmodels FAQ
What models does it cover?
OLS, GLM, discrete and mixed models, ARIMA and SARIMAX, with coefficient inference and diagnostics.
Does a successful fit prove the model is right?
No. The skill says a fit shows the procedure ran. Causal identification, calibrated uncertainty and model adequacy need a sound design and assumptions.
Which version was reviewed?
statsmodels 0.15.0, released on 27 August 2026.
What changed in version 0.15?
The skill advises result_object=True with named fields for ADF and KPSS, and rng= in place of seed or random_state.
How do I install it?
Run npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill statsmodels.
Is it checked for security?
Its security audits show passes from Gen Agent Trust Hub and Socket, and a warning from Snyk.
What are the best statsmodels alternatives?
Popular alternatives to statsmodels include aeon, pydeseq2 and pymc. See all statsmodels alternatives