Skip to main content

Table 4 Average number of non-zero coefficients and mean absolute error (MAE) of coefficient estimates over 100 simulations for scenario 1 and 2

From: Gsslasso Cox: a Bayesian hierarchical model for predicting survival and detecting associated genes by incorporating pathway information

 

Method

Average

Number

MAE

Scenario 1

gsslasso

8.61

0.60 (0.24)

lasso

51.99

3.77 (0.40)

grlasso

474.80

12.43 (1.64)

grMCP

62.00

9.30 (2.56)

grSCAD

108.80

8.41 (1.25)

cMCP

14.19

0.96 (0.34)

SGL

39.79

6.25 (1.65)

Scenario 2

gsslasso

9.74

1.29 (0.84)

lasso

53.70

4.02 (0.46)

grlasso

502.05

12.11 (2.08)

grMCP

57.13

8.04 (0.67)

grSCAD

167.59

8.77 (0.93)

cMCP

15.14

0.96 (0.33)

  1. *: the optimal s0 = 0.02 and s0 = 0.03 for gsslasso method under scenario 1 and 2, respectively. For scenarios with overlap structures, SGL method was not used for comparison since it cannot handle overlap situation directly