analyzing-regression-diagnostics

What problem does it solve?

This Skill checks whether a fitted linear or generalized linear regression is safe to interpret before you report coefficients, p-values, or confidence intervals, reducing the risk of misleading conclusions from violated assumptions.

Core Features & Use Cases

• Runs the standard diagnostic battery for linearity, residual normality, homoscedasticity, independence, leverage, Cook's distance, and multicollinearity.
• Explains what each failing diagnostic means and routes you to the right remediation, such as robust standard errors, transformation, refitting, or handling a justified influential point.
• Use it when you have an OLS, GLS, ridge, lasso, or GLM fit and need to determine whether the model is trustworthy for inference or coefficient interpretation.

Quick Start

Use this skill to review my fitted regression model, run the full diagnostic checklist, and tell me which assumption failures need remediation.