Skip to main content

Table 2 Prediction performance of various methods on Brier Score (mean (SD)) across six scenarios and 100 duplicated runs

From: A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data

 

Scenario 1

Scenario 2

Scenario 3

Scenario 4

Scenario 5

Scenario 6

Penalty and group penalty methods

Lasso

0.148 (0.010)

0.202 (0.009)

0.235 (0.006)

0.160 (0.009)

0.190 (0.010)

0.232 (0.006)

MCP

0.147 (0.011)

0.201 (0.010)

0.236 (0.007)

0.159 (0.010)

0.189 (0.012)

0.233 (0.007)

SCAD

0.146 (0.010)

0.201 (0.009)

0.236 (0.006)

0.159 (0.010)

0.189 (0.011)

0.232 (0.006)

network

0.165 (0.008)

0.202 (0.008)

0.237 (0.006)

0.156 (0.009)

0.193 (0.009)

0.232 (0.005)

gsslasso

0.145 (0.011)

0.201 (0.008)

0.235 (0.007)

0.156 (0.010)

0.188 (0.011)

0.232 (0.007)

grlasso

0.165 (0.008)

0.210 (0.006)

0.240 (0.005)

0.174 (0.008)

0.201 (0.007)

0.238 (0.005)

grMCP

0.179 (0.007)

0.213 (0.007)

0.240 (0.004)

0.184 (0.008)

0.205 (0.007)

0.238 (0.005)

grSCAD

0.168 (0.008)

0.210 (0.006)

0.240 (0.005)

0.176 (0.007)

0.201 (0.007)

0.238 (0.005)

cMCP

0.148 (0.012)

0.202 (0.010)

0.237 (0.008)

0.160 (0.012)

0.194 (0.011)

0.233 (0.007)

Model stacking methodsa

nsslasso (Lasso)

0.162 (0.009)

0.199 (0.008)

0.232 (0.007)

0.148 (0.010)

0.185 (0.010)

0.229 (0.007)

nsslasso (MCP)

0.159 (0.009)

–

–

–

–

0.229 (0.007)

nsslasso (SCAD)

0.159 (0.009)

–

–

–

–

0.228 (0.007)

nsslasso (network)

0.160 (0.008)

0.199 (0.008)

0.237 (0.006)

0.146 (0.010)

0.185 (0.010)

0.229 (0.007)

nsslasso (KNN)

0.190 (0.010)

0.219 (0.009)

0.244 (0.008)

0.179 (0.011)

0.208 (0.009)

0.240 (0.007)

nsslasso (NBa)

0.205 (0.009)

0.226 (0.008)

0.243 (0.005)

0.191 (0.009)

0.214 (0.008)

0.238 (0.005)

nsslasso (RF)

0.182 (0.009)

0.212 (0.008)

0.240 (0.006)

0.167 (0.011)

0.199 (0.009)

0.236 (0.006)

L-BFGS (Lasso)

0.176 (0.010)

0.206 (0.008)

0.234 (0.005)

0.167 (0.013)

0.196 (0.008)

0.232 (0.005)

nLasso (Lasso)

0.162 (0.009)

0.199 (0.008)

0.232 (0.006)

0.148 (0.010)

0.185 (0.010)

0.228 (0.006)

  1. aThe first column displays the super learner outside the bracket and the base learner inside. “–” means unanalyzed due to complexity in computation
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us