. Author manuscript; available in PMC: 2014 Mar 24.

Published in final edited form as: J Am Stat Assoc. 2013 Dec 19;108(504):1173–1188. doi: 10.1080/01621459.2013.810149

Table 5.

The optimal min-TMR rules under the exponential tilt assumption. Cut-off points were transformed back to the original scale of CD4 count. The numbers in parentheses are standard errors calculated using the bootstrap method.

cut-off points

FNR

FPR

TMR

lower

upper

S₁

77 (46)

.94 (.06)

.01 (.02)

.24 (.02)

.15

0 (29)

199 (10)

69 (.05)

.00 (.01)

.17 (.02)

.30

0 (10)

278 (11)

.47 (.04)

.00 (.00)

.11 (.01)

S_{1}^{*}

109 (23)

.89 (.05)

.03 (.01)

.24 (.02)

.15

57 (22)

206 (10)

.67 (.04)

.01 (.01)

.18 (.02)

.30

41 (17)

287 (12)

.46 (.04)

.01 (.00)

.12 (.01)