Lesson 9: Index Variables

February 29, 2024

Review:

• Indicator or “Dummy” variables for categorical data
• Used car data
• Singapore real estate transaction data

Presentation:

• Use MS Excel today
• Index Variables
  • Quantifying a categorical variable
  • Pivot tables make it easy
  • Demonstrate with this larger Pueblo real estate data set
  • Build a regression model with an Index variable

Activity:

• Build a Multiple Regression Model
  • Download recent Singapore real estate transaction data
    • All 2023 transactions
    • Apartment/Condo sales only
    • Use “Resale” only
    • I will assign 1+ postal zones to each student
  • Build a regression model to estimate selling prices
  • Create and Test Index Variables
  • Maximize R^2 and the F Statistic

Assignment:

• Team Modeling Competition
  • Combine all data sets from each student on your team
  • Build a new model on this larger combined data set
  • Maximize R^2 and the F Statistic
• Model Evaluation
  • Instructor will provide a set of properties for estimation
  • Estimates will be compared to actuals
  • Measure MAD and RMSE
  • Produce Actual vs Forecast scatter plot