EXAMPLE 10 Calculating and interpreting prediction errors (residuals)
Use the regression equation from Example 9 to calculate and interpret the prediction error (residual) for the following cities.
- Philadelphia: ,
- Washington, DC: ,
- The actual data point for Philadelphia is shown in the scatterplot in Figure 27 (denoted as “Phil”). In Example 9, we calculated the predicted high temperature for Philadelphia to be . In Figure 27, represents the -value of the point on the regression line where it intersects . That is, the actual high temperature lies directly above predicted temperature for low temperature .
- The actual high temperature in Washington that day was . Using the regression equation, the predicted high temperature is . So the prediction error is . The data point lies below the regression line, so that its actual high temperature of 65°F is lower than predicted given its low temperature of 45°F.
Philadelphia, Pennsylvania
Figure 4.27: FIGURE 27 Prediction error for Philadelphia and Washington, DC.