Fit the Bayesian model for Binary Replicates
BayesianFit(
ni,
si,
prior = list(a_FP = 2, b_FP = 2, a_FN = 2, b_FN = 2, a_T = 0.5, b_T = 0.5),
...
)An object of class stanfit returned by rstan::sampling.
The Stan model samples the posterior distribution of the fixed parameters \(p\), \(q\) and \(\theta\). It also generates the latent variables \(T_i\) according to their predictive distribution.
Numeric vector of \(n_i\)'s, the total numbers of replicates for each individual
Numeric vector of \(s_i\)'s, the numbers of replicates equal to 1 for each individual
A list of prior parameters for the model, see Details.
Arguments passed to rstan::sampling (e.g. iter, chains).
The model is a Bayesian model for binary replicates. The prior distribution is as follows:
The false positive rate: \(p \sim \text{Beta}(a_{FP}, b_{FP})\)
The false negative rate: \(q \sim \text{Beta}(a_{FN}, b_{FN})\)
The prevalence: \(\theta \sim \text{Beta}(a_T, b_T)\)
The statistical model considers that the true statuses are the latent $$T_i \sim \text{Bernoulli}(\theta).$$
And, given the true status \(T_i\), the number of positive replicates is $$S_i \sim \text{Binomial}(n_i, T_i(1-q)+(1-T_i)p).$$
credint, bayesian_scoring, classify_with_scores, bayesian_prevalence_estimate