Skip to main content

Table 6 Cross-validations of the EBLASSO-NE, EBLASSO-NEG and LASSO for the simulation with main and epistatic effects

From: Empirical Bayesian LASSO-logistic regression for multiple binary trait locus mapping

Algorithm

Parametersa

logL± STEb

 

0.0631

−0.44 ± 0.04

 

0.0891

−0.41 ± 0.04

 

0.1259

−0.39 ± 0.03

EBLASSO-NE

0.1600

−0.37 ± 0.01c

 

0.1778

−0.42 ± 0.04

 

0.2512

−0.53 ± 0.04

 

0.3548

−0.47 ± 0.04

 

(−0.4,0.05)

−0.40 ± 0.05

 

(−0.2,0.05)

−0.20 ± 0.05

 

(−0.1,0.05)

−0.10 ± 0.05

EBLASSO-NEG

(−0.2,0.01)

−0.35 ± 0.02

 

(−0.2,0.1)

−0.33 ± 0.02c

 

(−0.2,0.5)

−0.35 ± 0.02

 

0.1027

−0.57 ± 0.01

 

0.0511

−0.47 ± 0.02

LASSO

0.0254

−0.37 ± 0.02c

 

0.0127

−0.37 ± 0.03

 

0.0063

−0.39 ± 0.04

  1. aParameters are λ for EBLASSO-NE and LASSO, (a, b) for EBLASSO-NEG.
  2. bThe average log likelihood and standard error were obtained from ten-fold cross validation.
  3. cThe optimal log likelihood and corresponding parameter(s) chosen for comparison with other methods.
Back to article page

BMC Genomic Data

ISSN: 2730-6844

Contact us