Skip to main content

Table 3 Classification accuracy (%) for each imputation method using LDA, QDA, and BLR classifiers under both MAR and MCAR mechanisms

From: Classification of breast cancer recurrence based on imputed data: a simulation study

Missing data %

Imputation Method

Accuracy under (%) MAR

Accuracy under (%) MCAR

LDA

QDA

BLR

LDA

QDA

BLR

15%

SMEAN

70.14

69.12

70.42

71.10

68.20

71.00

HD

68.72

67.77

69.70

69.70

68.20

70.10

KNN

68.25

68.25

69.55

67.50

67.50

70.00

PMM

68.70

68.25

69.30

71.10

68.70

70.00

MICE

68.28

67.30

69.12

68.20

67.70

70.10

EMB

65.33

65.33

69.84

71.60

69.70

69.60

30%

SMEAN

70.10

68.20

71.40

72.90

72.00

70.80

HD

68.20

67.20

68.80

69.80

67.40

69.70

KNN

67.80

67.80

70.80

70.60

69.70

70.30

PMM

70.80

67.90

70.40

71.10

68.70

69.30

MICE

70.70

70.70

64.80

71.60

72.50

69.70

EMB

68.20

66.40

71.10

69.70

68.20

69.40

45%

SMEAN

73.50

73.90

73.40

73.90

72.00

72.40

HD

69.19

68.25

70.56

69.10

68.10

69.80

KNN

72.00

72.90

72.20

69.70

69.70

70.70

PMM

66.80

69.20

70.30

70.60

70.10

70.60

MICE

68.20

67.80

69.30

69.20

65.90

70.30

EMB

67.30

65.90

69.30

68.70

69.20

68.70

60%

SMEAN

70.10

70.10

72.30

72.00

69.70

71.70

HD

68.20

67.20

68.80

67.30

67.80

69.10

KNN

68.20

68.70

69.60

65.90

63.50

68.90

PMM

68.20

66.40

69.80

64.90

66.40

68.50

MICE

68.70

65.40

69.40

67.80

68.70

69.10

EMB

67.80

68.20

68.80

66.50

67.50

68.30