The Meaning of the Intercept
What You'll Learn
You'll understand what the intercept (β₀) represents in a linear regression model and how to interpret it in real-world contexts.
The Core Idea
In simple linear regression, our model is: Ŷ = β₀ + β₁X
The intercept (β₀) is the predicted value of Y when X equals zero. It's where your regression line crosses the Y-axis.
Think of it like a taxi fare: the intercept is your base fare before you've traveled any distance (X = 0 miles). If the fare is 2 per mile, that $3.50 is your intercept—what you pay just for getting in.
Practical Interpretation
When Zero Makes Sense
If you're predicting house prices from square footage, the intercept is the predicted price for a 0-square-foot house. That's nonsensical! The intercept often has no meaningful real-world interpretation when X = 0 is impossible or outside your data range.
When Zero Is Meaningful
If you're predicting test scores from hours studied, and X = 0 (no studying) is realistic, the intercept represents your expected "baseline" score—what you'd get without preparation.
Statistical Role
Even when β₀ lacks practical meaning, it's mathematically essential. It anchors your line at the correct vertical position to minimize prediction errors across all data points.
A Warning
Never extrapolate interpretation beyond your data range. If you measured houses from 800–3000 square feet, your intercept (at 0 sq ft) is a mathematical artifact, not a prediction.
Key Takeaway: The intercept is the predicted Y value when X = 0, but its practical meaning depends entirely on whether zero is a realistic, observed value in your context—it always serves a crucial mathematical role regardless.