Skip to main content

Table 4 Results of random effects expectation–maximization (REEM) tree algorithm for diastolic blood pressure (DBP)

From: A wide range of missing imputation approaches in longitudinal data: a simulation study and real data analysis

Imputation method

MSE

RMSE

MAD

Deviance

Complete cases

0.860

0.927

0.715

48402.31

Interpolation LOCF

0.853

0.924

0.698

155130.7

Interpolation global

0.858

0.926

0.672

162242.6

Interpolation local

0.864

0.929

0.654

167328.3

Interpolation bisector

0.859

0.927

0.669

162677.8

copyMean.LOCF

0.854

0.924

0.701

155243.2

copyMean.global

0.858

0.927

0.678

161725.4

copyMean.local

0.863

0.929

0.658

166834.6

copyMean.bisector

0.859

0.927

0.675

162305.9

LOCF

0.854

0.924

0.705

155906.2

NOCB

0.854

0.924

0.701

155904.5

Traj mean

0.851

0.922

0.658

152664.4

Traj median

0.851

0.923

0.662

152885.9

Traj hot deck

0.854

0.924

0.706

156429.8

Cross mean

0.871

0.933

0.70

155292

Cross median

0.871

0.933

0.70

155181.4

Cross hot deck

0.882

0.939

0.718

162979.3

FCS-LMM

0.881

0.939

0.721

161444.4

FCS-LMM-het

0.881

0.939

0.721

161444.4

FCS-GLMM

0.860

0.927

0.720

157721.4

FCS-LMM-LN

0.859

0.927

0.716

157916.3

FCS-LMM-LN-het

0.860

0.927

0.720

158532

JointAI

0.859

0.927

0.720

157743

hmi

0.859

0.927

0.720

157716

JM-SMC

0.859

0.927

0.720

157732

JM-SMC-het

0.859

0.927

0.720

157756.6

JM-MLMM

0.859

0.927

0.720

157715

JM-FJ

0.859

0.927

0.720

157638.5