Skip to main content

Table 2 Robustness of DAGBagM with respect to distributional assumptions

From: DAGBagM: learning directed acyclic graphs of mixed variables with an application to identify protein biomarkers for treatment response in ovarian cancer

Distribution

Power (TPR)

FDR

F1-score

Gaussian

0.8245

0.087

0.8665

t distribution (df = 3)

0.8375

0.105

0.8653

t distribution (df = 5)

0.8262

0.093

0.8647

Gamma distribution (shape = 1, scale = 2)

0.8182

0.1

0.8572

  1. The data are generated using a DAG with \(p = 102\) nodes and \(\parallel E\parallel =109\) edges. The range of signal-to-noise ratio (SNR) is set to be [0.5, 1.5], and the sample size is set to be \(n = 102\). The reported numbers are averaged over 100 independent replicates
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us