Relationships
Simple Linear Regression
Fit a straight line that predicts one variable from another.
Advertisement
Theory
Simple linear regression estimates an intercept and slope for the line ŷ = a + bx. The slope is the expected change in y for a one-unit increase in x.
ŷ = a + bx
What you will learn
- Estimate slope and intercept.
- Use a fitted line for prediction.
- Interpret residuals and model limitations.
How to interpret it
The slope estimates the average change in predicted y for a one-unit increase in x. The intercept is the predicted y when x equals zero, but it may have no practical meaning when zero lies outside the observed range.
Assumptions and limitations
Simple least-squares inference typically assumes a linear relationship, independent observations, roughly constant residual spread, and approximately normal residuals for tests and intervals. Predictions are most reliable within the observed x range.
Common mistakes
- Extrapolating far beyond the data.
- Treating the fitted line as proof of causation.
- Ignoring residual patterns that reveal nonlinearity or unequal variance.
Try it yourself
Problem question
For points (1,2), (2,4), and (3,5), estimate the regression line.
Step-by-step solution
- Compute x̄ = 2 and ȳ = 3.667.
- The slope is Σ(x−x̄)(y−ȳ) / Σ(x−x̄)² = 1.5.
- The intercept is 3.667 − 1.5(2) = 0.667.
- The fitted line is ŷ = 0.667 + 1.5x.
Calculator results are educational estimates. Confirm important academic or professional work independently.