Table 1 Parameter set simulation. For each parameter set realisation, true values for the statistical parameters were simulated by drawing from continuous uniform distributions, \(\mathcal {U}(\mathrm {min,\ max})\). Values for the design parameters were simulated by randomly drawing from fixed sets of values (in the case of sample sizes for simulated validation studies of a diagnostic test) or from a discrete uniform distribution (in the case of the sample size for an application of the test)
From: Comparison of Bayesian and frequentist methods for prevalence estimation under misclassification
Parameter | Description | Values sampled from |
|---|---|---|
Statistical | ||
\(\quad \quad \widetilde {{Se}\phantom {.}}\) | True sensitivity | \(\mathcal {U}\left (0.6,\ 1\right)\) |
\(\quad \quad \widetilde {{Sp}}\) | True specificity | \(\mathcal {U}\left (0.6,\ 1\right)\) |
\(\quad \quad \widetilde {\pi }\) | True prevalence | \(\mathcal {U}\left (0,\ 1\right)\) |
Design | ||
nSe | Sample size for a sensitivity validation study | {50,100,200,500,1000,2000, 5000} |
nSp | Sample size for a specificity validation study | {50,100,200,500,1000, 2000, 5000} |
n | Sample size for a test application | \(\mathcal {U}\left (50,\ 2000\right)\) |