For AG negative:(b)bb^ 0 ¼ 1 :963,bb^ 1 ¼ 0 :234,yy^¼ 9 :05;(c)t¼ 0 :987,
p¼ 0 :340;(d) r^2 ¼ 0 :065;(e)relationship is stronger and statistically
significant for AG positives but not so for AG negatives.8.10 (a)Term bb^ SEðbb^Þ tStatistic pValue
AG 2.847 1.343 2.119 0.043
Log(WBC) 0.234 0.243 0.961 0.345
AG-by-log(WBC) 0.583 0.321 1.817 0.080(b) Rather strong indication ðp¼ 0 : 080 Þ; (c) F¼7.69, df¼ð 3 ; 29 Þ,
p¼0.0006;(d)r^2 ¼ 0 :443.8.11 For revenue: (b) bb^ 0 ¼ð 1 : 113 Þð 10 ^3 Þ, bb^ 1 ¼ð 0 : 829 Þð 10 ^6 Þ, ^yy¼
ð 1 : 329 Þð 10 ^3 Þ;(c)t¼ 0 :198,p¼ 0 :844;(d)r^2 ¼ 0 :001.
For work load hours: (b) bb^ 0 ¼ð 0 : 260 Þð 10 ^3 Þ, bb^ 1 ¼ð 0 : 754 Þð 10 ^6 Þ,
^yy¼ð 1 : 252 Þð 10 ^3 Þ;(c)t¼ 1 :882,p¼ 0 :067;(d)r^2 ¼ 0 :078.
8.12 (a)Term bb^ SEðbb^Þ tStatistic pValue
Residency (3.176)(10^3 ) (1.393)(10^3 ) 2.279 0.028
Gender (0.348)(10^3 ) (0.303)(10^3 ) 1.149 0.258
Revenue (1.449)(10^6 ) (4.340)(10^6 ) 0.334 0.741
Hours (2.206)(10^6 ) (0.760)(10^6 ) 2.889 0.006
Residency by hours (2.264)(10^6 ) (0.930)(10^6 ) 2.436 0.020(b)Yes, rather strong indicationðp¼ 0 : 020 Þ;(c)F¼2.14, df¼ð 5 ; 38 Þ,
p¼0.083;(d)r^2 ¼ 0 :219.Chapter 11
11.8 Log-rank test:p¼ 0 :0896; generalized Wilcoxon test:p¼ 0 :1590.11.10 95% confidence interval for odds ratio:ð 1 : 997 ; 13 : 542 Þ; McNemar’s chi-
square:X^2 ¼ 14 :226;pvalue¼ 0 :00016.
11.11 McNemar’s chi-square:X^2 ¼ 0 :077;pvalue¼ 0 :78140.
11.12 95% confidence interval for odds ratio:ð 1 : 126 ; 5 : 309 Þ; McNemar’s chi-
square:X^2 ¼ 5 :452;pvalue¼ 0 :02122.
11.13 For men: McNemar’s chi-square,X^2 ¼ 13 :394;pvalue¼ 0 :00025; for
women: McNemar’s chi-square,X^2 ¼ 0 :439;pvalue¼ 0 :50761.
Chapter 12
12.2 ½ 0 : 93 þ 3 ð 0 : 1 Þð 0 : 9 Þ^2 ð 0 : 9 Þ^3 f 0 : 23 þ 3 ð 0 : 2 Þ^2 ð 0 : 8 Þþ 3 ð 0 : 2 Þð 0 : 8 Þ^2 ½ 1 ð 0 : 8 Þ^3 g
¼ 0 :264.
12.3 At each new dose level, enroll three patients; if no patient has DLT, the
trial continues with a new cohort at the next higher dose; if two or threeANSWERS TO SELECTED EXERCISES 525