TABLE 2

Results for different covariance structure (data driven) and different parameters of the model (7). For each multiple testing procedure, we provide the false discovery rate (FDR), marginal false discovery rate (mFDR), and expected true positives (TP) and their standard error. Based on 500 data generations. In bold, the most powerful procedure. |$K=15000,b=0,\tau = \sqrt{2}$|⁠.

Method|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH
ParametersAR(1), |$\pi=0.3,\rho=0.8$|Banded(1), |$\pi=0.3,\rho=0.5$|
mFDR0.03620.04250.03770.04370.03550.04200.03530.03800.04220.04320.03510.0415
s.e.(mFDR)0.00070.00030.00020.00070.00070.00070.00050.00040.00030.00060.00050.0005
FDR0.03580.04250.03770.04320.03510.04150.03520.03800.04220.04300.03480.0413
s.e.(FDR)0.00070.00030.00020.00070.00060.00070.00050.00040.00030.00050.00050.0005
TP241.81100.31274.4273.7242.2269.6241.0532.0621.7273.2241.8269.6
s.e.(TP)1.16561.63912.09121.18891.11641.23581.02671.34191.29431.15361.10971.1918
ParametersAR(1), |$\pi =0.1$|⁠, |$\rho = 0.8$|Banded(1), |$\pi =0.1$|⁠, |$\rho = 0.5$|
mFDR0.04610.04910.04380.05420.04710.04950.04570.04250.04670.05490.04790.0507
s.e.(mFDR)0.00180.00190.00180.00180.00170.00190.00150.00080.00070.00150.00150.0015
FDR0.04350.04890.04390.05160.04410.04630.04510.04230.04660.05380.04620.0490
s.e.(FDR)0.00170.00190.00190.00180.00170.00170.00150.00070.00060.00150.00150.0015
TP41.3353.5394.846.543.745.241.5137.4188.146.343.344.7
s.e.(TP)0.45151.23781.58190.47140.48800.49800.38780.63670.67700.44030.45610.4606
Method|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH
ParametersAR(1), |$\pi=0.3,\rho=0.8$|Banded(1), |$\pi=0.3,\rho=0.5$|
mFDR0.03620.04250.03770.04370.03550.04200.03530.03800.04220.04320.03510.0415
s.e.(mFDR)0.00070.00030.00020.00070.00070.00070.00050.00040.00030.00060.00050.0005
FDR0.03580.04250.03770.04320.03510.04150.03520.03800.04220.04300.03480.0413
s.e.(FDR)0.00070.00030.00020.00070.00060.00070.00050.00040.00030.00050.00050.0005
TP241.81100.31274.4273.7242.2269.6241.0532.0621.7273.2241.8269.6
s.e.(TP)1.16561.63912.09121.18891.11641.23581.02671.34191.29431.15361.10971.1918
ParametersAR(1), |$\pi =0.1$|⁠, |$\rho = 0.8$|Banded(1), |$\pi =0.1$|⁠, |$\rho = 0.5$|
mFDR0.04610.04910.04380.05420.04710.04950.04570.04250.04670.05490.04790.0507
s.e.(mFDR)0.00180.00190.00180.00180.00170.00190.00150.00080.00070.00150.00150.0015
FDR0.04350.04890.04390.05160.04410.04630.04510.04230.04660.05380.04620.0490
s.e.(FDR)0.00170.00190.00190.00180.00170.00170.00150.00070.00060.00150.00150.0015
TP41.3353.5394.846.543.745.241.5137.4188.146.343.344.7
s.e.(TP)0.45151.23781.58190.47140.48800.49800.38780.63670.67700.44030.45610.4606
TABLE 2

Results for different covariance structure (data driven) and different parameters of the model (7). For each multiple testing procedure, we provide the false discovery rate (FDR), marginal false discovery rate (mFDR), and expected true positives (TP) and their standard error. Based on 500 data generations. In bold, the most powerful procedure. |$K=15000,b=0,\tau = \sqrt{2}$|⁠.

Method|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH
ParametersAR(1), |$\pi=0.3,\rho=0.8$|Banded(1), |$\pi=0.3,\rho=0.5$|
mFDR0.03620.04250.03770.04370.03550.04200.03530.03800.04220.04320.03510.0415
s.e.(mFDR)0.00070.00030.00020.00070.00070.00070.00050.00040.00030.00060.00050.0005
FDR0.03580.04250.03770.04320.03510.04150.03520.03800.04220.04300.03480.0413
s.e.(FDR)0.00070.00030.00020.00070.00060.00070.00050.00040.00030.00050.00050.0005
TP241.81100.31274.4273.7242.2269.6241.0532.0621.7273.2241.8269.6
s.e.(TP)1.16561.63912.09121.18891.11641.23581.02671.34191.29431.15361.10971.1918
ParametersAR(1), |$\pi =0.1$|⁠, |$\rho = 0.8$|Banded(1), |$\pi =0.1$|⁠, |$\rho = 0.5$|
mFDR0.04610.04910.04380.05420.04710.04950.04570.04250.04670.05490.04790.0507
s.e.(mFDR)0.00180.00190.00180.00180.00170.00190.00150.00080.00070.00150.00150.0015
FDR0.04350.04890.04390.05160.04410.04630.04510.04230.04660.05380.04620.0490
s.e.(FDR)0.00170.00190.00190.00180.00170.00170.00150.00070.00060.00150.00150.0015
TP41.3353.5394.846.543.745.241.5137.4188.146.343.344.7
s.e.(TP)0.45151.23781.58190.47140.48800.49800.38780.63670.67700.44030.45610.4606
Method|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH|$\hat{\boldsymbol{T}}_0$|-rule|$\hat{\boldsymbol{T}}_1$|-rule|$\hat{\boldsymbol{T}}_2$|-ruleSun and CaiBHABH
ParametersAR(1), |$\pi=0.3,\rho=0.8$|Banded(1), |$\pi=0.3,\rho=0.5$|
mFDR0.03620.04250.03770.04370.03550.04200.03530.03800.04220.04320.03510.0415
s.e.(mFDR)0.00070.00030.00020.00070.00070.00070.00050.00040.00030.00060.00050.0005
FDR0.03580.04250.03770.04320.03510.04150.03520.03800.04220.04300.03480.0413
s.e.(FDR)0.00070.00030.00020.00070.00060.00070.00050.00040.00030.00050.00050.0005
TP241.81100.31274.4273.7242.2269.6241.0532.0621.7273.2241.8269.6
s.e.(TP)1.16561.63912.09121.18891.11641.23581.02671.34191.29431.15361.10971.1918
ParametersAR(1), |$\pi =0.1$|⁠, |$\rho = 0.8$|Banded(1), |$\pi =0.1$|⁠, |$\rho = 0.5$|
mFDR0.04610.04910.04380.05420.04710.04950.04570.04250.04670.05490.04790.0507
s.e.(mFDR)0.00180.00190.00180.00180.00170.00190.00150.00080.00070.00150.00150.0015
FDR0.04350.04890.04390.05160.04410.04630.04510.04230.04660.05380.04620.0490
s.e.(FDR)0.00170.00190.00190.00180.00170.00170.00150.00070.00060.00150.00150.0015
TP41.3353.5394.846.543.745.241.5137.4188.146.343.344.7
s.e.(TP)0.45151.23781.58190.47140.48800.49800.38780.63670.67700.44030.45610.4606
Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close