# Table 4 Simulation results based on 1000 data sets with sample size of 50 -- Per cent censoring (40%, 25%)

Per cent censoring True ρ True ρ c Method Mean $ρ ^ c$ Relative bias (%) Empirical SD Mean SE The percentage of 95% confidence intervals that include true value of CCC
(40%, 25%) 0.25 0.238 1. Delete the pair 0.1334 −43.95 0.1858 0.1652 88.1
2. Replace by LOD 0.1908 −19.83 0.1525 0.1150 84.3
3. Replace by 0.5 × LOD 0.1796 −24.54 0.1467 0.1155 80.2
4. Replace by c × LOD 0.1780 −25.22 0.1528 0.1152 81.8
5. ML 0.2296 −3.53 0.1395 0.1345 92.4
0.50 0.476 1. Delete the pair 0.2735 −42.54 0.1693 0.1516 76.9
2. Replace by LOD 0.3673 −22.84 0.2110 0.1004 80.2
3. Replace by 0.5 × LOD 0.3366 −29.29 0.2771 0.1011 74.3
4. Replace by c × LOD 0.3537 −25.70 0.1788 0.1011 75.9
5. ML 0.4617 −3.00 0.1206 0.1135 93.2
0.75 0.714 1. Delete the pair 0.4913 −31.19 0.1373 0.1184 62.3
2. Replace by LOD 0.6022 −15.66 0.1705 0.0696 78.7
3. Replace by 0.5 × LOD 0.5706 −20.08 0.2039 0.0710 74.7
4. Replace by c × LOD 0.5689 −20.32 0.2151 0.0713 74.1
5. ML 0.7003 −1.92 0.0805 0.0750 93.2
1. Abbreviations: LLD Lower Limit of Detection, ML Maximum Likelihood, C = Random number from uniform distribution (0, 1), CI confidence interval, Relative bias = $Mean ρ ^ c − True ρ c True ρ c ×100%$, CCC5= $ρ c = 2 COV X , Y VAR X + VAR Y + E X − E Y 2$. 