Skip to main content

Table 5 Percentage in decrease of mean power

From: Re-sampling strategy to improve the estimation of number of null hypotheses in FDR control under strong correlation structures

 

ρ

True π0

BYa

SampG.qb

SampS.qc

SampG.md

SampS.me

Data-B

0.3

0.9

48.57%

4.45%

3.93%

3.38%

3.54%

  

0.8

41.43%

1.96%

2.06%

1.58%

1.79%

  

0.7

36.81%

1.89%

2.01%

2.30%

2.42%

  

0.6

34.73%

1.58%

1.63%

2.00%

2.18%

 

0.5

0.9

18.21%

0.47%

0.52%

0.40%

0.51%

  

0.8

11.02%

0.28%

0.46%

0.16%

0.34%

  

0.7

10.55%

0.26%

0.36%

0.21%

0.28%

  

0.6

10.38%

0.33%

0.57%

0.36%

0.53%

 

0.7

0.9

5.67%

0.34%

0.52%

0.38%

0.54%

  

0.8

5.73%

0.83%

1.21%

0.98%

1.15%

  

0.7

7.91%

0.14%

0.49%

0.09%

0.31%

  

0.6

4.40%

0.16%

0.41%

0.21%

0.35%

Data-M

 

0.9

11.85%

0.36%

0.53%

0.27%

0.53%

  

0.8

9.50%

0.19%

0.36%

0.17%

0.36%

  

0.7

7.99%

0.09%

0.23%

0.01%

0.12%

  

0.6

7.76%

0.08%

0.19%

0.02%

0.15%

  1. a (Powerqvalue - PowerBY)/Powerqvalue
  2. b (Powerqvalue - PowerSampG.q)/Powerqvalue
  3. c (Powerqvalue - PowerSampS.q)/Powerqvalue
  4. e (Powerqvalue - PowerSampG.m)/Powerqvalue
  5. f (Powerqvalue - PowerSampS.m)/Powerqvalue