statsmodels
statsmodels provides classes and functions for estimating statistical models, running hypothesis tests, and exploring data in Python.
A Python library for statistical modeling, hypothesis testing, regression, ANOVA, diagnostics, and data exploration.
About the Project
Why use this?
statsmodels provides classes and functions for estimating many statistical models, conducting statistical tests, and exploring data. It supports R-style formulas and pandas DataFrames, making it accessible for engineers analyzing process data, experiments, and quality metrics.
Lean and Six Sigma practitioners choose statsmodels for the statistical core of DMAIC projects: regression, ANOVA, hypothesis testing, confidence intervals, diagnostics, time-series analysis, and multivariate methods. Its focus on statistical inference and diagnostics makes it especially appropriate when an engineer needs to explain and defend a decision statistically rather than merely produce a predictive model. [399][402][412]
Key Features
- Statistical model estimation, hypothesis testing, and data exploration
- Linear and generalized linear models, ANOVA, and regression diagnostics
- R-style formula interface and pandas DataFrame support
- Extensive result statistics and model diagnostics
- Strong fit for DMAIC, process capability, and quality analytics
- Reference / Repo
- https://github.com/statsmodels/statsmodels
- Tool / Hosted
- https://www.statsmodels.org/
- Replaces / Alternative to
- Minitab, JMP
- Free layer
- open-source