Skip to main content
. Author manuscript; available in PMC: 2013 May 1.
Published in final edited form as: Stat. 2012 Nov 23;1(1):103–114. doi: 10.1002/sta.411

Table 2.

Results for the second simulation scenario using 1000 Monte Carlo data sets. E{Y*(gopt)} = 8.12. Entries as Table 1.

n=200 n=500 n=1000

Estimator Ê(ĝopt) Ê(ĝopt) Ê(ĝopt) Ê(ĝopt) Ê(ĝopt) Ê(ĝopt)
PS correct

RG 7.82 (0.32) 7.78 (0.13) 7.84 (0.21) 7.83 (0.10) 7.85 (0.16) 7.84 (0.08)
Zhang et al. 8.09 (0.32) 8.01 (0.08) 8.11 (0.20) 8.07 (0.03) 8.11 (0.15) 8.09 (0.02)
CAIPWE 7.96 (0.33) 7.68 (0.13) 8.03 (0.20) 7.83 (0.06) 8.04 (0.15) 7.89 (0.04)
ĈIPWE 7.11 (0.41) 6.96 (0.31) 7.39 (0.28) 7.28 (0.23) 7.55 (0.22) 7.47 (0.17)
Ĉreg 7.67 (0.34) 7.57 (0.17) 7.73 (0.21) 7.68 (0.11) 7.73 (0.16) 7.70 (0.08)

PS incorrect

RG 7.79 (0.30) 7.78 (0.13) 7.82 (0.19) 7.83 (0.10) 7.82 (0.14) 7.84 (0.08)
Zhang et al. 8.01 (0.30) 8.02 (0.06) 8.03 (0.18) 8.07 (0.03) 8.03 (0.14) 8.09 (0.02)
ĈAIPWE 7.87 (0.30) 7.70 (0.11) 7.95 (0.18) 7.85 (0.05) 7.96 (0.13) 7.89 (0.04)
ĈIPWE 7.11 (0.37) 7.09 (0.26) 7.22 (0.24) 7.19 (0.16) 7.25 (0.18) 7.22 (0.12)
Ĉreg 7.63 (0.31) 7.57 (0.17) 7.70 (0.20) 7.68 (0.11) 7.70 (0.15) 7.70 (0.08)