Skip to main content

Table 2 Simulation results

From: Introducing a new estimator and test for the weighted all-cause hazard ratio

Sc.

Assumptions for

Assumptions for

τ

Weights

Ln of

Mean number

Mean of estimated

Power for

 

original parametric

new non-parametric

   

True

of events (sd)

Ln(WHR) (sd)

permutation

weight-based

 

estimator*

estimator#

   

WHR

    

test

log-rank test

    

\(w_{EP_{1}}\)

\(w_{EP_{2}}\)

\(ln(\theta _{CE}^{w}(\tau))\)

E P 1

E P 2

\(ln(\hat {\theta }^{w}_{CE}(\tau))\)

\(ln(\tilde {\theta }^{w}_{CE}(\tau))\)

\(\hat {\theta }^{w}_{CE}(\tau)\)

\(\tilde {\theta }^{w}_{CE}(\tau)\)

1a

\(\checkmark \)

1

1

0.1

-0.60

50.21 (6.51)

35.16 (5.21)

-0.61 (0.27)

-0.57 (0.28)

0.66

0.72

1b

  

2

1

0.1

-0.67

72.60 (6.74)

80.13 (6.73)

-0.67 (0.20)

-0.60 (0.24)

0.94

0.90

1c

  

2

0.1

1

-1.18

  

-1.20 (0.22)

-1.14 (0.21)

1.00

1.00

2a

1

1

0.1

-0.44

48.22 (6.08)

37.15 (5.33)

-0.59 (0.29)

-0.58 (0.29)

0.58

0.75

2b

  

2

1

0.1

-0.38

67.79 (6.75)

51.93 (5.60)

-0.54 (0.23)

-0.53 (0.23)

0.64

0.81

3a

\(\checkmark \)

1

1

0.1

-0.88

39.97 (5.63)

47.94 (5.87)

-0.90 (0.29)

-1.00 (0.31)

0.90

0.96

3b

  

1

0.1

1

0.43

  

0.42 (0.29)

0.38 (0.28)

0.00

0.00

3c

  

2

1

0.1

-0.68

69.84 (6.39)

124.63 (6.40)

-0.67 (0.20)

-0.80 (0.21)

0.96

1.00

4a

1

1

0.1

-0.39

62.84 (6.39)

6.49 (2.50)

-0.22 (0.64)

-0.23 (0.26)

0.14

0.29

4b

  

1

0.1

1

-0.08

  

0.12 (0.96)

0.01 (0.44)

0.02

0.05

4c

  

2

1

0.1

-0.46

86.66 (6.95)

24.29 (4.57)

-0.27 (0.21)

-0.28 (0.21)

0.27

0.44

4d

  

2

0.1

1

-0.36

  

-0.12 (0.40)

-0.15 (0.32)

0.05

0.17

5a

1

1

0.1

-0.56

133.27 (6.26)

19.31 (4.05)

-0.82 (0.28)

-0.82 (0.17)

1.00

1.00

5b

  

1

0.1

1

-0.03

  

-0.04 (0.57)

-0.14 (0.30)

0.03

0.34

6a

\(\checkmark \)

2

1

0.1

0.00

67.04 (6.74)

56.49 (6.37)

0.00 (0.21)

0.00 (0.23)

0.02

0.08

6b

  

2

0.1

1

0.00

  

-0.00 (0.25)

-0.00 (0.24)

0.02

0.06

7a

\(\checkmark \)

\(\checkmark \)

2

1

0.1

0.00

15.83 (3.84)

141.89 (6.18)

0.01 (0.30)

0.02 (0.28)

0.02

0.02

7b

  

2

0.1

1

0.68

  

0.68 (0.16)

0.68 (0.17)

0.00

0.00

8a

\(\checkmark \)

2

1

0.1

-0.56

108.25 (6.93)

78.79 (6.69)

-0.69 (0.21)

-0.56 (0.19)

0.92

0.96

8b

  

2

0.1

1

-1.12

  

-1.29 (0.24)

-1.13 (0.21)

1.00

1.00

9a

1

0.1

1

-0.88

132.78 (6.43)

15.22 (3.56)

-1.10 (0.42)

-0.93 (0.31)

0.77

0.97

9b

  

2

1

0.1

-0.51

178.24 (4.21)

21.76 (4.21)

-0.66 (0.18)

-0.52 (0.15)

0.96

0.97

9c

  

2

0.1

1

-0.71

  

-1.31 (0.48)

-0.89 (0.30)

0.90

0.99

10a

1

1

0.1

-0.64

84.95 (7.02)

62.44 (6.34)

-0.58 (0.21)

-0.59 (0.21)

0.81

0.94

10b

  

1

0.1

1

-0.95

  

-1.04 (0.23)

-1.05 (0.23)

1.00

1.00

10c

  

2

1

0.1

-0.72

113.64 (6.92)

80.58 (6.55)

-0.55 (0.17)

-0.61 (0.18)

0.89

0.98

10d

  

2

0.1

1

-0.79

  

-0.95 (0.19)

-1.03 (0.21)

1.00

1.00

  1. Ln: natural logarithm; WHR: Weighted all-cause hazard ratio; sd: Standard deviation;
  2. *It is assumed that the Weibull model used to estimate the cause-specific hazards is the correct one;
  3. #It is assumed that the cause-specific baseline hazards are equal.