Lesson 7: Categorical Data and Indicator Variables

October 13, 2025

Review:

• Multicollinearity
• Interaction Variables

Presentation:

• Pivot tables to aggregate by category
• Indicator Variables (aka “dummy” variables)
  • incorporate categorical variables
  • binary – value is either 1 or 0
  • if 0, then the coefficient has no effect on y-hat
  • if 1, then the coefficient is added to the estimate
• Demonstration with Sample Real Estate data
  • Use a Pivot table to summarize price by neighborhood
  • create a dummy variable for a neighborhood
  • create a dummy variable for multiple neighborhoods

Activity:

• Use the Pueblo Real Estate model (from last lesson)
  • Incorporate one or more Indicator variables
  • Can you improve R^2 and/or F with an indicator variable?

Assignment:

• Download Used Cars data and utilize Excel (dataset is large enough to slow Sheets)
  • Create a correlation matrix for all numeric variables.
  • Do you see evidence of multicollinearity?
  • If so, incorporate one or more interaction variables
  • Incorporate one or more Indicator variables to include categorical data elements (e.g., Make, Model, Trim or Type)
  • After experimenting with Interaction and Indicator variables, make a model with whatever variables you think work best, to estimate used vehicle sales prices.