Computational and Mathematical Methods in Medicine

Research Article

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

Table 1

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


		0.2			0.5			0.8
	Method	No. selected variables	TPR	FPR	No. selected variables	TPR	FPR	No. selected variables	TPR	FPR

200	ALASSO	36.913 (5.119)	0.816 (0.161)	0.007 (0.005)	33.682 (2.255)	0.969 (0.075)	0.006 (0.004)	26.242 (1.796)	0.988 (0.047)	0.004 (0.003)
	AENET	36.532 (4.503)	0.830 (0.174)	0.006 (0.004)	33.571 (1.909)	0.958 (0.089)	0.006 (0.004)	26.636 (1.808)	0.989 (0.046)	0.004 (0.004)
	SCAD	37.684 (4.484)	0.858 (0.162)	0.007 (0.003)	35.414 (2.691)	0.972 (0.073)	0.006 (0.006)	25.054 (2.136)	0.993 (0.042)	0.004 (0.005)
	SCAD-L₂	37.190 (3.171)	0.867 (0.156)	0.006 (0.003)	35.960 (2.742)	0.960 (0.092)	0.006 (0.006)	25.935 (2.232)	0.994 (0.034)	0.004 (0.005)
	MCP	27.401 (2.996)	0.860 (0.142)	0.004 (0.002)	25.125 (2.308)	0.975 (0.068)	0.004 (0.005)	23.634 (1.889)	0.994 (0.034)	0.004 (0.004)
	MCP-L₂	27.030 (3.103)	0.849 (0.149)	0.004 (0.003)	25.881 (2.115)	0.971 (0.079)	0.004 (0.004)	23.328 (1.643)	0.995 (0.031)	0.004 (0.004)
	Boosting (Binder)	41.750 (4.874)	0.899 (0.140)	0.007 (0.005)	39.940 (4.890)	0.987 (0.052)	0.007 (0.005)	38.272 (4.731)	1.000 (0.000)	0.007 (0.006)
	Oracle	5.000	1.000	0.000	5.000	1.000	0.000	5.000	1.0000	0.000

400	ALASSO	34.871 (1.430)	0.963 (0.820)	0.006 (0.005)	31.846 (1.196)	1.000 (0.000)	0.005 (0.003)	24.572 (0.891)	1.000 (0.000)	0.004 (0.002)
	AENET	34.181 (1.649)	0.972 (0.075)	0.006 (0.003)	31.701 (1.041)	0.999 (0.014)	0.005 (0.002)	24.631 (0.925)	1.000 (0.000)	0.004 (0.002)
	SCAD	35.192 (1.821)	0.987 (0.053)	0.006 (0.003)	29.942 (1.679)	1.000 (0.000)	0.004 (0.004)	25.304 (1.260)	1.000 (0.000)	0.004 (0.003)
	SCAD-L₂	35.736 (1.805)	0.980 (0.067)	0.006 (0.005)	29.672 (1.470)	0.998 (0.020)	0.004 (0.003)	25.150 (1.164)	1.00 (0.000)	0.004 (0.003)
	MCP	24.140 (1.580)	0.968 (0.073)	0.004 (0.003)	20.761 (1.227)	0.999 (0.014)	0.003 (0.002)	18.572 (1.228)	1.000 (0.000)	0.003 (0.002)
	MCP-L₂	24.162 (1.468)	0.972 (0.072)	0.004 (0.003)	20.661 (1.105)	0.998 (0.020)	0.003 (0.002)	18.381 (0.916)	0.999 (0.014)	0.003 (0.002)
	Boosting (Binder)	39.911 (5.256)	0.995 (0.031)	0.007 (0.005)	39.280 (4.874)	1.000 (0.000)	0.007 (0.005)	38.695 (4.965)	1.000 (0.000)	0.007 (0.004)
	Oracle	5.000	1.000	0.000	5.000	1.000	0.000	5.000	1.00	0.00

TPR: true positive rate; FPR: false positive rate; : sample size.