baker: Bayesian Analysis Kit for Etiology Research can be accessed at https://github.com/zhenkewu/baker.
Description of baker
baker is motivated by the Pneumonia Etiology Research for Child Health (PERCH) study which seeks to understand disease etiology using case-control data from multiple sources with measurement errors. baker is designed to catalyze effective communication between research analysts and practicing clinicians. If you are interested in estimating the population etiology pie (fraction) for a disease or the probability of each cause for an individual case, try baker. The main function is `nplcm()` which fits the model with or without covariates.
baker implements hierarchical Bayesian models to infer disease etiology for multivariate binary data. The package builds in functionalities for data tidying and exploration, model specification, estimation, visualization, diagnostics, and comparisons. baker can implement models for dependent multivariate binary measurements given disease status, regression analyses of etiology, multiple imperfect measurements, covariate-stratified priors for true positive rates among cases, and multiple-pathogen etiology.