Table 2. .
The FDR for testing 3,000 random entries for N = 50 and 5,000 random entries for N = 200 are averaged over 5,000 iterations. These random subsets retain the sparsity in the corresponding edge correlation matrix. The significance level α sets to be {0.05, 0.1, 0.2}, and sample size T ranges from 400 to 700.
| N | α | T = 400 | T = 450 | T = 500 | T = 550 | T = 600 | T = 650 | T = 700 |
|---|---|---|---|---|---|---|---|---|
| N = 50 | α = 0.05 | 0.017 | 0.012 | 0.007 | 0.005 | 0.004 | 0.004 | 0.005 |
| α = 0.1 | 0.020 | 0.015 | 0.011 | 0.009 | 0.009 | 0.009 | 0.011 | |
| α = 0.2 | 0.027 | 0.023 | 0.019 | 0.017 | 0.018 | 0.019 | 0.021 | |
| N = 200 | α = 0.05 | 0.057 | 0.035 | 0.019 | 0.010 | 0.008 | 0.006 | 0.005 |
| α = 0.1 | 0.060 | 0.035 | 0.022 | 0.015 | 0.012 | 0.009 | 0.009 | |
| α = 0.2 | 0.066 | 0.039 | 0.028 | 0.021 | 0.019 | 0.016 | 0.017 |