Table 5 The bias and RMSE in parentheses of the Rogan-Gladen adjusted prevalence estimates when targeting a specificity of 0.95. The method(s) with the smallest bias in each scenario or equivalent after rounding are bolded
Scenario A | Scenario B | |||||||
|---|---|---|---|---|---|---|---|---|
n=50 | n=200 | n=50 | n=200 | |||||
Prevalence=0.05 | ||||||||
Empirical | 0.01 | (0.0323) | 0.00 | (0.0173) | 0.02 | (0.0293) | 0.00 | (0.0152) |
Normal | 0.02 | (0.0287) | 0.02 | (0.0186) | 0.01 | (0.0404) | -0.01 | (0.0260) |
Log Normal | 0.03 | (0.0292) | 0.02 | (0.0242) | 0.03 | (0.0369) | 0.02 | (0.0236) |
MAD | 0.10 | (0.1049) | 0.10 | (0.0997) | 0.16 | (0.1558) | 0.15 | (0.1536) |
Log MAD | 0.06 | (0.0621) | 0.06 | (0.0566) | 0.11 | (0.1129) | 0.11 | (0.1114) |
Pareto 0.9 | 0.01 | (0.0263) | 0.00 | (0.0146) | 0.00 | (0.0298) | -0.02 | (0.0215) |
Hybrid Empirical | 0.01 | (0.0324) | 0.00 | (0.0173) | 0.02 | (0.0301) | 0.00 | (0.0152) |
Hybrid Pareto 0.9 | 0.01 | (0.0286) | 0.00 | (0.0147) | 0.00 | (0.0308) | -0.02 | (0.0215) |
Prevalence=0.30 | ||||||||
Empirical | 0.01 | (0.0281) | 0.00 | (0.0153) | 0.01 | (0.0223) | 0.00 | (0.0122) |
Normal | 0.02 | (0.0245) | 0.01 | (0.0153) | 0.01 | (0.0295) | -0.01 | (0.0196) |
Log Normal | 0.02 | (0.0239) | 0.02 | (0.0186) | 0.02 | (0.0280) | 0.01 | (0.0179) |
MAD | 0.08 | (0.0781) | 0.07 | (0.0732) | 0.11 | (0.1146) | 0.11 | (0.1133) |
Log MAD | 0.05 | (0.0468) | 0.04 | (0.0415) | 0.08 | (0.0830) | 0.08 | (0.0822) |
Pareto 0.9 | 0.01 | (0.0244) | 0.00 | (0.0135) | 0.00 | (0.0225) | -0.01 | (0.0164) |
Hybrid Empirical | 0.01 | (0.0279) | 0.00 | (0.0153) | 0.01 | (0.0229) | 0.00 | (0.0122) |
Hybrid Pareto 0.9 | 0.01 | (0.0256) | 0.00 | (0.0135) | 0.00 | (0.0233) | -0.01 | (0.0164) |