The Multiple Regression Equation
What you'll learn: How to write and interpret the mathematical form of multiple linear regression with multiple predictors.
The Core Idea
In simple linear regression, you predicted an outcome using one predictor. Multiple linear regression extends this by letting you use several predictors simultaneously. The equation looks like this:
Y = β₀ + β₁X₁ + β₂X₂ + ... + βₚXₚ + ε
Let's break down each piece:
- Y: The outcome variable you're predicting (also called the dependent variable or response)
- β₀: The intercept — the expected value of Y when all X variables equal zero
- β₁, β₂, ..., βₚ: The slope coefficients — each tells you how much Y changes when that specific X increases by 1 unit, holding all other predictors constant
- X₁, X₂, ..., Xₚ: Your predictor variables (independent variables)
- p: The number of predictors in your model
- ε (epsilon): The error term — the difference between actual Y and what the model predicts
Interpreting the Equation
Think of this like a recipe where each ingredient (predictor) contributes to the final dish (outcome). Each β coefficient tells you how much that ingredient matters, independent of the others.
For example, predicting house price:
Price = β₀ + β₁(SquareFeet) + β₂(Bedrooms) + β₃(Age) + ε
If β₁ = 150, it means each additional square foot adds $150 to the price, even after accounting for bedrooms and age.
Key Takeaway: The multiple regression equation Y = β₀ + β₁X₁ + β₂X₂ + ... + βₚXₚ + ε lets you model how multiple predictors simultaneously influence an outcome, with each coefficient representing the unique contribution of its predictor while holding others constant.