BMC Medical Research Methodology

Table 3 Relative bias and Mean Square Error (MSE) of the cut-point in the normal homoscedastic scenario ^† with diseased and disease-free fractions of 50% and a censoring level of 38%

From: Optimal cut-point definition in biomarkers: the case of censored failure time outcome

				Youden index		Concordance probability		Point closest-to-(0,1) corner
J(c_opt)^‡	CZ(c_opt)^‡	c_opt	N	Relative bias	MSE	Relative bias	MSE	Relative bias	MSE
0.2	0.36	0.25	50	0.1345	0.3419	0.1918	0.1514	0.1812	0.1099
			100	0.1347	0.2592	0.1042	0.0894	0.1225	0.0627
			200	0.0547	0.2084	0.0638	0.0572	0.0652	0.0413
			400	0.0932	0.1395	0.0177	0.0350	0.0200	0.0245
0.4	0.49	0.52	50	0.0912	0.2147	0.0829	0.1303	0.0624	0.0891
			100	0.0595	0.1418	0.0460	0.0841	0.0438	0.0518
			200	0.0106	0.1039	0.0082	0.0541	0.0024	0.0339
			400	0.0085	0.0641	−0.0021	0.0314	0.0009	0.0173
0.6	0.64	0.84	50	0.0446	0.1596	0.0560	0.1262	0.0507	0.0890
			100	0.0628	0.1094	0.0599	0.0794	0.0492	0.0494
			200	0.0258	0.0653	0.0256	0.0492	0.0212	0.0260
			400	0.0042	0.0408	0.0052	0.0287	0.0012	0.0160
0.8	0.81	1.28	50	0.0461	0.1404	0.0435	0.1340	0.0310	0.1186
			100	0.0449	0.0883	0.0411	0.0792	0.0282	0.0587
			200	0.0132	0.0503	0.0108	0.0462	0.0073	0.0297
			400	0.0039	0.0367	0.0041	0.0323	0.0030	0.0185

^†X_Z ≤ τ ~ N(μ_Z ≤ τ, 1), X_Z >τ ~ N(0, 1). ^‡The levels of J and CZ are achieved by μ_Z ≤ τ = 0.51, 1.05, 1.68, 2.56, respectively.

Back to article page

ISSN: 1471-2288

Contact us

General enquiries: journalsubmissions@springernature.com