Light

residual

Definition

A residual is the difference between an observed value and its predicted value in a regression model. It represents how much the actual data point deviates from the estimated line or curve.

Analogy

Think of residuals as measuring how far off you are from your intended destination during a road trip. If you arrive exactly where you planned, the residual is zero. But if you end up somewhere else, the residual indicates how much you missed your target.

Related terms

Least Squares Regression Line: The least squares regression line is a line that minimizes the sum of squared residuals. It provides the best fit for a set of data points.

Homoscedasticity: Homoscedasticity refers to when the variability of residuals is constant across all levels of an independent variable in a regression model.

Standardized Residuals: Standardized residuals are calculated by dividing each residual by its standard deviation. They help identify outliers and assess whether assumptions of linear regression are met.