Multiple Linear Regression
Adding variables, interaction terms, and interpreting richer models
In simple linear regression we used a single variable — salary — to estimate market value. But we can also build better models for estimating market value if we can add more variables that are useful for explaining market value.
Multiple linear regression lets us include more than one predictor (independent variable). We'll start with an additive model (parallel lines for different groups), then add an interaction term to allow the relationship to differ between groups.
We'll look at what happens if we add another independent variable to the model – the player's position.
Adding a Second Variable
From one predictor to two — the additive model
Interpreting the Coefficients
What each number in the equation actually means
Interaction Terms
When the effect of salary depends on position
Comparing Models
R², adjusted R², and choosing the right model
A Continuous Second Variable
What changes when both predictors are numbers, not categories
Interactions with Continuous Variables
When the effect of salary depends on age — interaction terms and simple slopes
References & Further Reading
Books, articles, and resources to deepen your understanding