Skip to main content

Table 2 Estimated AUC (95% confidence interval) by elastic net shrinkage; training dataset from [8], test dataset from [30]

From: Quasi-linear score for capturing heterogeneous structure in biomarkers

 

LL

QL (K=2)

ε=0.25

0.732 (0.665, 0.796)

0.755 (0.691, 0.814)

ε=0.50

0.723 (0.655, 0.788)

0.754 (0.691, 0.813)

ε=0.75

0.707 (0.636, 0.776)

0.748 (0.684, 0.807)

  1. A parameter ε denotes the proportion of ridge regularization to lasso regularization
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us