Here€™s a plot of the Studentized residuals from the regression model of Exercise 18 plotted against ArterialMPH. The plot is colored according to City Size (Small, Medium, Large, and Very Large), and regression lines are fit for each city size.

a) The model in Exercise 18 includes indicators for city size. Considering this display, have these indicator variables accomplished what is needed for the regression model? Explain.
Here is another model that adds two new constructed variables to the model in Exercise 18. They are the product of Arterial MPH and Small and the product of Arterial MPH and VeryLarge.

b) What does the predictor AM*Sml (Arterial MPH by Small) do in this model? Interpret the coefficient.
c) Does this model improve on the model in Exercise 18? Explain.

Transcribed Image Text:

small = medium e- large very large 125 0.00T 1.25 32 30 Arterial mph 26 Dependent variable s: Delay/person R-squared= 80.7% R-squared (adjusted) 78.5% s 6.316 with 68 8 60 degrees of freedom Source Regression 10013.0 Residual 2393.82 Sum of Squares df Mean Square F-ratio 7 1430.44 60 39.8970 35.9 Variable Coeff Intercept HiWayMPH ArterialMPH -2.60848 0.69 Smal1 Large VeryLarge AM* Sml AM*VLg SE(Coeff) t-ratio P-value 8.79 <0.0001 -4.21 <0.0001 0.0004 0.0646 0.0202 0.1623 0.0712 0.1491 153.110 17.42 -1.02104 0.2426 67 -3.74 2.39 1.84 -125.979 66.92 4.89837 2.053 -89.4993 63.25 3.81461 2.077 3.38139 2.314 1.46