. Author manuscript; available in PMC: 2018 Jan 1.

Published in final edited form as: J R Stat Soc Ser C Appl Stat. 2016 Jun 27;66(2):313–328. doi: 10.1111/rssc.12164

Table 3.

Estimation and testing of α₁ (truth is 0.5): light censoring dependent on Y.

	Method	Bias	SD	SE	MSE	Type 1 error	Power
n = 100	No Censoring	0.0060	0.2709	0.2624	0.0689	0.052	0.524
	Complete Case	0.0484	0.3702	0.3619	0.1333	0.050	0.348
	Single Imputation¹	−0.0104	0.3199	0.3016
	Multiple Imputation	0.0087	0.3003	0.3199	0.0709	0.048	0.366
	Deletion Thresholding	−0.0375	0.2306	0.2342	0.0563	0.055	0.119
	Complete Thresholding	−0.0343	0.2117	0.2151	0.0474	0.052	0.102
	Reverse Survival²					0.048	0.416

n = 250	No Censoring	0.0001	0.1605	0.1605	0.0258	0.047	0.872
	Complete Case	0.0368	0.2278	0.2222	0.0507	0.051	0.690
	Single Imputation¹	−0.0174	0.3301	0.1819
	Multiple Imputation	0.0222	0.2144	0.2103	0.0447	0.046	0.722
	Deletion Thresholding	0.1278	0.1544	0.1539	0.0400	0.048	0.377
	Complete Thresholding	1.2149	0.1335	0.1352	1.4938	0.046	0.251
	Reverse Survival²					0.049	0.754

: MSE, type I error and power not reported due to invalid SE for single imputation

: bias, SD and SE not reported due to non-interpretability of estimate from reverse survival regression