In Problem 1, separate straight-line regressions of SBP on QUET were compared for smokers (SMK = 1) and nonsmokers (SMK = 0).
a. Define a single multiple regression model that uses the data for both smokers and nonsmokers and that defines straight-line models for each group with possibly differing intercepts and slopes. Define the intercept and slope for each straight-line model in terms of the regression coefficients of the single regression model.
b. Using the accompanying computer output, determine and plot on graph paper the two fitted straight lines obtained from the fit of the regression model.
c. Test the following null hypotheses:
H0: "The two lines are parallel."
H0: "The two lines are coincident."
For each of these tests, state the appropriate null hypothesis in terms of the regression coefficients of the regression model.
d. Compare your answers in parts (b) and (c) with those you obtained by fitting separate regressions in Problem 1.

Transcribed Image Text:

Regression of SBP on QUET, SMK, and QUET × SMK ANALYSIS OF VARIANCE Sum of DF Squares Mean Square F Value Pr>F 1394.70253 Source Model Error Corrected Total 1742 0001 4184.10759 28 2241.86116 31 6425 96875 80.06647 Root MSE Dependent Mean Coeff Vair 8.94799 144.53125 6.19104 R-Square 0.6511 Adj R-Sq 0 6137 PARAMETER ESTIMATES Parameter Standard Variable Intercept QUET SMK QUETSMK Error t Value Pr>t Type I SS 19,97235 DFEstimate 247 00199 668457 61 $.0001 3537,94574 129.94357 24.16355 24 02256 582 42075 1-6.18478 6.93171-0.89 0.3799 63.74110 49.31176 26.30283 5,703494