1. The SAS dataset “112211” on blackboard gives the systolic blood pressure (SBP), body size (QUET), age (AGE), and smoking history (SMK = 0 if a nonsmoker, SMK = 1 if a current or previous smoker) for a hypothetical sample of 32 white males over 40 years old from the town of Angina. (QUET stands for “quetelet index,” a measure of size defined by QUET = 100 (weight/height2).)
a. On the accompanying scatter plot for QUET vs. AGE, sketch by eye a line that fits the data reasonably well. Comment on the relationship described.
b. (1)
Determine the least-squares estimates of the slope ( 1) and intercept ( 0) for the
straight-line regression of QUET (Y) on AGE (X).
. (2) Sketch the estimated regression line on the scatter diagram in part (a) involving QUET and AGE. Compare this new line with the line you drew in part (a).
. (3) Test the null hypothesis of zero slope; be sure to interpret the result.
. (4) Based on your test in part (b)(3), would you conclude that body size increases
as age increases? Explain your answer.
. (5) Obtain a 95% confidence interval for 1. In1terpret your result.
. (6) Would you reject the null hypothesis H0: = 0 in part (b)(3) based on the confidence
interval that you calculated in part (b)(5)? Explain. (7) Use SAS to find and draw 95% confidence bands and prediction bands with fitted
line on the scatter plot. (Must enable html output option in SAS: Tools->Options ->Preferences->Results->Check “Create HTML”.)
1
(8) Find the 99% confidence interval and prediction interval for an individual with AGE = 45 from SAS output. Interpret your answer. (9) Determine r and r2, and interpret your results, respectively. (10) Test H0: ρ = 0 versus H1: ρ ≠ 0 at α = .01, respectively. (Calculate by hand).