Skip to main content

Table 4 Comparison of TPR, LTPR, and FPR across 300 simulated datasets, each with five causal regions

From: Resampling-based tests for Lasso in genome-wide association studies

Model

TPR(t=0.3)

LTPR(t=0.3)

FPR(t=0.3)

TPR(t=0.5)

LTPR(t=0.5)

FPR(t=0.5)

Lasso-PL

0.780

0.166

0.0087

0.762

0.232

0.0091

Lasso-Ayers

0.775

0.166

0.0092

0.755

0.232

0.0096

Lasso-AL

0.761

0.159

0.0075

0.741

0.222

0.0079

SMA

0.753

0.336

0.0101

0.740

0.448

0.0115

  1. Each dataset contains n=600 subjects, 1000 SNPs and five causal regions. All testing was done using significance level α=0.01. See Section “Simulation 2” for a definition of “causal region”, TPR, LTPR, and FPR