Table 5.

Simulation results showing the mean values of the point estimators and the corresponding standard errors (SE) under different assumed values of θ. There were 10 trial replicates for each value of θ. The probability of stopping at stage 1 was 0.15, 0.37 and 0.65 for θ = 0.10, 0.14 and 0.18, respectively.

Type of estimator	Estimator	Difference in proportions (SE)
		θ = 0.10	θ = 0.14	θ = 0.18
MLE/naive	MLE (overall)	0.103 (0.054)	0.144 (0.054)	0.184 (0.053)
Unconditionally unbiased/bias-adjusted	MLE (stage 1)	0.100 (0.057)	0.140 (0.057)	0.180 (0.057)
	Median unbiased estimator (MUE)	0.101 (0.053)	0.142 (0.054)	0.182 (0.054)
	UMVUE	0.100 (0.052)	0.140 (0.054)	0.180 (0.055)
	Bias-corrected MLE (UBC-MLE)	0.101 (0.054)	0.142 (0.055)	0.183 (0.054)
Conditionally unbiased/bias-adjusted	MLE (stage 2)	0.100 (0.111)	0.140 (0.111)	0.180 (0.111)
	Conditional MUE (CMUE)	0.115 (0.083)	0.152 (0.080)	0.190 (0.081)
	UMVCUE	0.100 (0.062)	0.140 (0.071)	0.179 (0.080)
	Bias-corrected MLE (CBC-MLE)	0.111 (0.067)	0.154 (0.073)	0.194 (0.078)