Regularized Weighted Nonparametric Likelihood Approach for High-Dimension Sparse Subdistribution Hazards Model for Competing Risk Data

Table 3

Results of the simulation studies for proportional odds model () with 5 informative variables () for and scenario. Values shown are means (standard deviations) of each performance measure over 500 replicates (: ~40% average censoring; ).

							No. selected variables					FDR
	No. selected variables	TPR	FDR	No. selected variables	TPR	FDR		TPR	FDR	No. selected variables	TPR
	ALASSO	37.233 (3.143)	0.972 (0.073)	0.006 (0.003)	32.445 (2.341)	1.000 (0.000)		0.005 (0.004)	38.113 (2.027)	0.903 (0.087)	0.006 (0.004)		33.271 (2.271)	0.996 (0.033)	0.006 (0.003)
AENET	37.523 (2.881)	0.970 (0.077)	0.006 (0.005)	32.611 (2.254)	1.000 (0.000)	0.005 (0.002)	35.792 (1.745)	0.902 (0.124)	0.006 (0.003)	34.215 (2.421)	0.995 (0.041)	0.006 (0.003)
SCAD	31.231 (2.779)	0.978 (0.067)	0.005 (0.004)	30.472 (2.471)	1.000 (0.000)	0.005 (0.002)	36.723 (2.717)	0.901 (0.108)	0.006 (0.005)	32.344 (2.622)	0.982 (0.031)	0.005 (0.003)
SCAD-L2	32.485 (2.812)	0.979 (0.063)	0.005 (0.005)	30.285 (2.345)	0.999 (0.002)	0.005 (0.002)	36.785 (3.299)	0.903 (0.120)	0.006 (0.005)	33.426 (2.312)	0.990 (0.027)	0.006 (0.003)
MCP	27.131 (3.110)	0.980 (0.061)	0.004 (0.005)	24.634 (2.331)	1.000 (0.000)	0.004 (0.002)	25.872 (1.983)	0.907 (0.124)	0.004 (0.004)	25.121 (2.107)	0.990 (0.033)	0.004 (0.003)
MCP-L2	27.743 (3.103)	0.982 (0.060)	0.004 (0.005)	24.411 (2.262)	1.000 (0.000)	0.004 (0.002)	25.921 (2.146)	0.912 (0.137)	0.004 (0.004)	25.423 (1.998)	0.996 (0.024)	0.004 (0.003)