Table 4.

The first two columns show the pooled estimated coefficients and standard errors (SE) from the logistic regressions on $\mathrm{Pr}\left(I_{Y_5<0.1}\right)$ (n = 300) using the full datasets (FD). The remaining columns show the average relative differences (%) as compared to those from FD for the complete case analysis (CC), log-normal imputation (LMlog), predictive mean matching (PMM), and symmetric (QRTs) and asymmetric (QRTa) transformed quantile regression imputation models. The latter were fitted with either (1) known or (2) unknown λ_p.

	FD		CC		LMlog		PMM		QRTs (1)		QRTa (1)		QRTs (2)		QRTa (2)
	Est.	SE	Est.	SE	Est.	SE	Est.	SE	Est.	SE	Est.	SE	Est.	SE	Est.	SE
Model 1
θ0	3.3	0.6	1.0	24.0	2.3	22.0	0.0	21.1	0.2	21.2	−5.3	21.1	2.9	21.7	−3.4	21.7
θ1	−1.8	0.3	1.0	16.6	0.6	14.9	0.2	14.7	0.4	14.8	−4.0	14.0	2.0	15.2	−2.8	14.4
θ2	−1.8	0.3	1.3	14.6	−3.0	12.6	−0.0	13.4	0.4	13.5	−4.5	12.8	1.1	13.9	−4.1	13.0
θ3	−0.2	0.2	0.1	14.6	−2.6	13.9	−1.2	14.3	−1.4	13.7	−5.6	12.8	−0.3	14.1	−5.8	13.3
Model 2
θ0	1.1	0.4	−14.1	23.6	−6.5	22.7	6.4	27.4	−0.5	23.1	−13.8	22.8	1.1	23.5	−13.6	22.7
θ1	−0.6	0.2	−8.7	13.9	−4.3	13.5	4.9	16.4	0.4	14.7	−8.0	14.7	1.5	15.0	−7.8	14.4
θ2	−0.8	0.2	−7.6	9.5	−10.5	9.8	−8.0	10.0	−0.4	11.5	−4.9	11.0	−0.1	11.1	−5.0	10.9
θ3	−0.1	0.1	−10.2	10.3	−16.1	10.4	−12.2	11.5	−3.8	12.5	−8.9	11.5	−4.8	11.7	−9.7	11.9
Model 3
θ0	−1.5	0.3	0.4	49.5	−24.9	38.4	5.2	97.1	2.6	55.8	2.8	55.8	3.0	54.7	2.5	55.7
θ1	0.2	0.3	1.2	34.5	−159.2	28.5	29.6	70.5	15.2	39.3	16.6	39.3	17.8	38.3	14.5	39.2