Bayesian Estimation of the True Prevalence and of the Diagnostic Test Sensitivity and Specificity of Enteropathogenic Yersinia in Finnish Pig Serum Samples

Figure 1

Graphical model used for the Bayesian analyses to estimate the true prevalence, presenting the conditional dependency structure between variables. The observed variable, , is the number of pigs detected positive in the sample of size . The priors are beta distributions for the sensitivity and specificity of the diagnostic test. is the true prevalence, for a truly positive farm. Since a farm can be truly nonpositive, the actual true prevalence is , where represents an indicator variable that a farm is truly positive.