Agent skill · data analytics · aj-geddes
Regression Modeling
Build predictive models using linear regression, polynomial regression, and regularized regression for continuous prediction, trend forecasting, and relationship quantification
Why this skill is useful
Provides specific implementation examples and metrics for various regression techniques that the AI may not generate reliably on its own.
What it needs
Requires matplotlib, numpy, pandas, scikit-learn, seaborn installed locally. About 4k tokens when loaded. Last updated 2026-03-04. 307 stars on the source repository.
What this skill does
Regression Modeling Overview Regression modeling predicts continuous target values based on input features, establishing quantitative relationships between variables for forecasting and analysis. When to Use Predicting sales, prices, or other continuous numerical outcomes Understanding relationships between independent and dependent variables Forecasting trends based on historical data Quantifying the impact of features on a target variable Building baseline models for comparison with more complex algorithms Identifying which variables most influence predictions Regression Types Linear Regression: Straight-line fit to data Polynomial Regression: Non-linear relationships Ridge (L2): Regularization to prevent overfitting Lasso (L1): Feature selection through regularization ElasticNet: Combines Ridge and Lasso Robust Regression: Resistant to outliers Key Metrics R² Score: Proportion of variance explained RMSE: Root Mean Squared Error MAE: Mean Absolute Error AIC/BIC: Model comparison criteria Implementation with Python Assumption Checking Linearity: Relationship is linear Independence: Observations are independent Homoscedasticity: Constant variance of errors Normality: Errors are normally distributed No multicollinearity: Features not highly correlated Model Selection Simple data: Linear regression Non-linear patterns: Polynomial regression Many features: Lasso or ElasticNet Outliers: Robust regression Prevent overfitting: Ridge or ElasticNet Deliverables Fitted models with coefficients R² and RMSE metrics Residual plots and analysis Cross-validation results Regularization parameter tuning curves Model comparison summary Predictions with confidence intervals
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aj-geddes/regression-modeling