Skip to main content

Table 3 Simulation results for estimation of incident/dynamic cause-specific AUC of a single marker in Scenario 2(b), comparing the methods- cWMR & cFPL

From: Inference about time-dependent prognostic accuracy measures in the presence of competing risks

   

cWMR

cFPL

Cause

Log(t)

AUC

Mean

ARB

BSE

SE

CP(%)

Mean

ARB

BSE

SE

CP(%)

1

-1.5

0.843

0.843

0.0

0.022

0.025

92.0

0.843

0.0

0.017

0.018

93.3

 

-1.2

0.827

0.830

0.31

0.023

0.026

94.0

0.829

0.245

0.018

0.02

92.3

 

-0.9

0.807

0.809

0.26

0.025

0.028

93.8

0.810

0.042

0.020

0.021

89.9

 

-0.6

0.784

0.786

0.25

0.029

0.031

94.8

0.788

0.451

0.025

0.026

93.6

 

-0.3

0.760

0.765

0.62

0.036

0.038

94.0

0.761

0.162

0.033

0.033

92.9

 

0.0

0.737

0.747

1.41

0.048

0.049

90.8

0.737

0.016

0.054

0.046

92.7

 

0.3

0.716

0.727

1.60

0.073

0.70

88.0

0.728

1.66

0.096

0.085

89.7

 

0.6

0.696

0.710

2.07

0.131

0.110

83.9

0.692

0.46

0.218

0.179

84.6

2

-1.5

0.5

0.5029

0.58

0.049

0.052

88.9

0.5028

0.56

0.048

0.0517

92.3

 

-1.2

0.5

0.5017

0.34

0.043

0.046

90.0

0.5025

0.50

0.042

0.45

94.3

 

-0.9

0.5

0.5004

0.08

0.040

0.043

89.9

0.5003

0.06

0.040

0.044

93.7

 

-0.6

0.5

0.4987

0.26

0.042

0.044

88.7

0.4983

0.34

0.045

0.044

92.1

 

-0.3

0.5

0.5005

0.10

0.046

0.049

93.1

0.4991

0.178

0.052

0.054

93.5

 

0.0

0.5

0.5006

0.12

0.057

0.06

90.5

0.5030

0.60

0.080

0.07

92.5

 

0.3

0.5

0.5043

0.86

0.082

0.081

89.3

0.5039

0.78

0.146

0.126

95.1

 

0.6

0.5

0.5042

0.84

0.142

0.124

87.5

0.4938

0.62

0.276

0.259

85.0

  1. Average of bootstrap mean (Mean), average of bootstrap standard errors (BSE), absolute relative bias (ARB(%)), model-based standard error (SE) and coverage probability (CP(%))-(nominal level —90%) of AUC