Skip to main content

Table 3 Summary of simulation results for different sizes of each block, when ρ = 0.6 and block = 1 under a normal scenario. nIMS and nTMS stand for the number of the true informative markers selected and the total number of markers selected, respectively.

From: AucPR: An AUC-based approach using penalized regression for disease prediction with high-dimensional omics data

Size

Method

Auc

Sensitivity

nIMS

nTMS

5

AucEN

0.84

0.50

3

4

 

AucL

0.82

0.45

3

4

 

LogEN

0.82

0.45

2

2

 

LogL

0.80

0.40

1

1

 

MSauc

0.81

0.40

1

4

40

AucEN

0.86

0.55

20

24

 

AucL

0.86

0.55

13

18

 

LogEN

0.85

0.50

4

4

 

LogL

0.82

0.45

2

2

 

MSauc

0.81

0.45

1

3