Table 5:

Gaussian data with n = 200 and p = 51 under MAR. Approximately 40% features and 92% cases contain missing values.

Models	Style	Time(s)	Imp MSE	Bias(${\widehat{\beta }}_1$)	CR(${\widehat{\beta }}_1$)	SE(${\widehat{\beta }}_1$)	SD(${\widehat{\beta }}_1$)
SoftImpute	SI	2.7	0.0132	−0.0017	0.92	0.0623	0.0642
GAIN	SI	35.9	1.356	0.3213	0.38	0.1142	0.4262
MIWAE	SI	46.5	0.0361	−0.0238	0.90	0.0632	0.0738
Linear RR	SI	628.4	0.1712	0.0358	0.91	0.1287	0.1568
MICE	MI	2.1	0.0200	0.0031	0.97	0.0644	0.0567
Sinkhorn	MI	42.1	0.1081	−0.1225	0.60	0.0978	0.1269
MI-NNGP1	MI	3.4	0.0129	0.0048	0.95	0.0621	0.0647
MI-NNGP1-BS	MI	4.8	0.0177	0.0052	0.97	0.0794	0.0624
MI-NNGP2	MI	5.5	0.0092	0.0083	0.96	0.0639	0.0563
MI-NNGP2-BS	MI	13.5	0.0105	0.0083	0.98	0.0705	0.0574
Complete data	-	-	-	0.0025	0.98	0.0605	0.0524
Complete case	-	-	-	0.1869	0.79	0.2298	0.2419
ColMean Imp	SI	-	0.4716	0.5312	0.28	0.2242	0.1729