Skip to main content
. Author manuscript; available in PMC: 2017 Apr 1.
Published in final edited form as: Ann Stat. 2016 Apr;44(2):629–659. doi: 10.1214/15-AOS1380

Table 6.

Results of the Real Data Example. “NZ”, #0.05, and #0.10 stand for the numbers of parameters that are nonzero, that has a p-value greater or equal to 0.05, and that has a p-value greater or equal to 0.1. “R2” denotes the R-squared value. “AIC”, “BIC” and “Obj.” stand for Akaike’s information criterion, Bayesian information criterion and objective function value.

Method SCAD: λ = 0.02; a = 3.7
NZ #0.05 #0.10 R2 AIC BIC Obj.
GM1 1500 — — 0.997 357.101 6563.691 212.279
GM2 401 401 401 0.698 246.886 1906.114 103.626
LLA0 185 119 115 0.864 −554.093 211.387 76.031
LLA1 181 83 80 0.912 −763.718 14.789 71.673
MIPGO 129 35 34 0.898 −796.581 −262.814 68.474
MCP: λ = 0.03; a = 2
GM1 818 — — 0.332 1448.966 5129.436 169.091
GM2 134 110 104 0.735 −296.624 −346.474 93.870
LLA0 96 5 6 0.856 704.654 −307.432 72.645
LLA1 113 2 2 0.902 −849.842 −382.279 69.292
MIPGO 109 3 3 0.899 −841.280 −390.267 68.591