Research Article

Evaluation of Four Multiple Imputation Methods for Handling Missing Binary Outcome Data in the Presence of an Interaction between a Dummy and a Continuous Variable

Table 4

Comparison of the performance of imputation methods based on the model based SE of regression coefficients over 1000 simulations.

VariableMethodMARMCAR
10%20%30%40%50%10%20%30%40%50%

(BMIreal)MCR0.1610.1670.1740.1830.1900.1610.1690.1740.1820.190
MI0.1640.1760.1900.2090.2330.1650.1770.1910.2080.230
MRF0.1630.1710.1780.1850.1890.1640.1710.1780.1840.192
MS0.1490.1460.1460.1460.1490.1490.1460.1460.1460.148
(Betotal)MCR0.1140.1180.1220.1240.1280.1150.1190.1220.1250.128
MI0.1170.1250.1350.1470.1630.1180.1260.1360.1480.163
MRF0.1160.1200.1250.1290.1330.1170.1210.1250.1280.131
MS0.1050.1020.1010.1010.1020.1050.1020.1010.1010.102
(PACS4)MCR0.1050.1090.1120.1130.1170.1050.1090.1120.1140.118
MI0.1080.1180.1260.1400.1540.1080.1150.1250.1360.147
MRF0.1070.1110.1140.1170.1190.1060.1110.1140.1180.120
MS0.0950.0930.0920.0920.0930.0960.0930.0920.0920.093
(PSPS)MCR0.1050.1090.1120.1140.1160.1050.1080.1110.1140.117
MI0.1090.1170.1270.1400.1540.1080.1150.1240.1350.149
MRF0.1070.1110.1150.1170.1190.1060.1100.1130.1170.120
MS0.0950.0930.0920.0920.0930.0960.0930.0920.0920.093
(Gender)MCR0.1920.1980.2040.2100.2150.1920.1990.2040.2100.215
MI0.1970.2100.2280.2450.2700.1980.2120.2280.2490.274
MRF0.1940.2020.2080.2160.2190.1950.2040.2090.2170.222
MS0.1780.1750.1740.1750.1780.1790.1750.1740.1760.178
(BMIreal ∗Gender)MCR0.2380.2460.2540.2640.2700.2390.2490.2550.2610.268
MI0.2430.2610.2810.3070.3390.2430.2610.2800.3050.342
MRF0.2410.2480.2530.2590.2600.2410.2490.2540.2580.263
MS0.2090.1990.1920.1890.1890.2100.1990.1930.1890.188