This R script uses the classic Ames Housing dataset to illustrate a full regression modeling project.
- Create Target Population
- Treat Missing Values
- Correlation
- Outlier Treatment
- Deep Dive into 3 Highly Correlated Variables
- Train/Test Split
- Automated Feature Selection (Forward, Backward, Stepwise)
- Model Results (VIF, R-squared, AIC, BIC, MSE, RMSE, MAE)
- Validation
- Reduce Model Complexity
- Diagnostic Analysis