Lesson 6: Interaction Variables in Multiple Regression
October 6, 2025
Review:
- Regression Output Quiz
- Career Event Friday
Presentation:
- MLB Baseball model
- 2013 MLB Team statistics
- Walk through model development
- Here’s a video where I do the same modeling work in Excel
- Applied Regression Analysis
- Demonstrate with Sample Real Estate data
- Problem: Multicollinearity
- Solution: Interaction Variables
- combine separate variables mathematically
- add, multiply, etc
- another way to avoid problems with multicollinearity
- Estimate selling price for a 1500 sqft, 3 bd, 2 ba home
Activity:
- Pueblo Real Estate with Interaction
- Use this larger Pueblo real estate data set
- Build a regression model to estimate selling price
- You must use 1 interaction variable
- Experiment with multiple combinations of variables to find the best model, maximizing R^2 and the F Statistic
- Use your regression model to estimate selling prices for these 4 hypothetical properties
- 3 BR, 2 BA, 1600 SqFt
- 4 BR, 3 BA, 2400 SqFt
- 3 BR, 1 BA, 1200 SqFt
- 2 BR, 1 BA, 800 SqFt
Assignment:
- MLB rebuild
- Watch this video where I build a MLB model in Excel
- Rebuild your 2013 MLB model in Excel (or if you’ve been working in Excel, rebuild in Sheets)
- Repeat with your 2025 MLB data
- Attempt to use an Interaction variable with the MLB data (either 2013 or 2025 model)
