Using the second measurement of the systolic pressure of the hepatic artery
as our dependent variable, the resulting ANOVA table is shown in Table 8.8.
The result of the overallFtest (p¼ 0 :0002) indicates that taken collectively,
the three independent variables (systolic pressure at transplant, gender, and
age) contribute significantly to the prediction of the dependent variable. In
addition, we have the results shown in Table 8.9. The e¤ects of pressure at
transplant and age are significant at the 5% level, whereas the e¤ect of gender is
not (p¼ 0 :5982).
Note:An SAS program would include these instructions:
DATA;
INPUT POST PRE SEX AGE;
DATALINES;
46 35 2 16
40 40 2 19
...
35 27 1 9
25 31 1 9
;
PROC REG;
MODEL POST = PRE SEX AGE;
which gives us all of the results above.
Example 8.8 There have been times the city of London experienced periods of
dense fog. Table 8.10 shows such data for a very severe 15-day period which
included the number of deaths in each day (y), the mean atmospheric smoke
(x 1 , in mg/m^3 ), and the mean atmospheric sulfur dioxide content (x 2 , in ppm).
TABLE 8.8
Source of Variation SS df MS FStatistic pValue
Regression 1810.93 3 603.64 12.158 0.0002
Error 844.02 17 49.65
Total 2654.95 20
TABLE 8.9
Variable Coe‰cient Standard Error tStatistic pValue
Pressure at transplant 0.381 0.082 4.631 0.0002
Gender 1.740 3.241 0.537 0.5982
Age 0.935 0.395 2.366 0.0301
300 CORRELATION AND REGRESSION