A library to check OLS assumptions.
Project description
OlsCheck
olscheck is a Python library designed to check and visualize the assumptions of Ordinary Least Squares (OLS). This tool streamlines the process of model diagnostics, offering an all-in-one solution for analyzing residuals, leverage, and multi-collinearity.
Features
- Residuals vs Fitted Plot: Visualize potential non-linear patterns.
- QQ-Plot: Check the normality of residuals.
- Scale-Location Plot: Confirm homoscedasticity.
- Leverage Plot: Identify influential cases.
- VIF Test: Assess multi-collinearity among predictors.
Background
Much of the code for this library is inspired by Statsmodels. However, the Statsmodels implementation is tightly coupled with its own components, making it challenging to use if you are employing OLS from a different library.
OlsCheck provides a straightforward alternative for generating key diagnostic plots that assess the assumptions underlying Ordinary Least Squares (OLS) regression. The library internally calculates residuals, leverage, and studentized residuals to deliver these insights.
Installation
pip install olscheck
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
