Adaptive L_1/2 Shooting Regularization Method for Survival Analysis Using Gene Expression Data

Table 1

The simulation results based on the high dimensional simulated dataset by the three methods over 100 replications. The columns include the average number of the selected variable (Var), the average number of the correct zeros (Corr), the average number of the incorrect zeros (Incorr), and the integrated Brier score (IBS). (Lasso: the Lasso method, A-L: the adaptive Lasso method, and L_1/2: the adaptive L_1/2 shooting regularization method).

𝑛	25% censoring					40% censoring
	Method	Var	Corr (994)	Incorr (0)	IBS	Var	Corr (994)	Incorr (0)	IBS
200	Lasso	81.29	917.29	0.26	0.1502	96.38	906.83	0.31	0.1516
	A-L	41.06	962.47	0.35	0.1474	59.05	948.89	0.43	0.1503
	L1/2	17.79	984.28	0.42	0.1440	20.42	974.15	0.53	0.1498
250	Lasso	98.46	903.07	0.11	0.1462	148.87	883.85	0.15	0.1493
	A-L	64.10	949.46	0.17	0.1446	74.42	933.74	0.26	0.1478
	L1/2	27.38	972.95	0.25	0.1421	31.91	968.03	0.34	0.1458
300	Lasso	167.82	883.18	0.01	0.1448	177.50	869.83	0.03	0.1479
	A-L	72.95	932.49	0.02	0.1436	80.97	927.42	0.06	0.1459
	L1/2	33.45	967.12	0.03	0.1418	38.64	958.38	0.06	0.1427
350	Lasso	196.24	847.84	0.00	0.1441	204.22	834.53	0.00	0.1463
	A-L	82.80	928.07	0.00	0.1428	89.18	921.54	0.00	0.1441
	L1/2	37.58	959.78	0.00	0.1405	40.15	948.63	0.00	0.1412