upmfit: upmfit: Bayesian MCMC sampling with the Unified Probability...
In upmfit: Unified Probability Model Fitting

The upmfit package uses methods described by McIntosh, et al. "Extensions to Bayesian Generalized Linear Mixed Effects Models for Household Tuberculosis Transmission," Statistics in Medicine (2017), to implement functions that generate model script in the JAGS language for the Unified Probability Model (UPM), and to run a JAGS MCMC sampler for posterior probability desnity estimation with the model script.

It should be noted that, while the UPM and R package upmfit were designed specifically with tuberculosis household contact studies in mind, the method is applicable to any scenario where there are two competing risks for a single outcome with unobserved linkage between the sources of risk and the outcome. Example situation where this method could be applicable are in profiling the risk of MRSA infection from a nosocomial versus other source, or in modeling risk of TB infection among populations utilizing homeless shelters. Non-clinical examples could be in an industrial process control setting where a manufactured item has two sources of degradation, one observed and one unobserved, or in a chemical mixing procedure where catalysis can occur from the mixing of an exogenous agent or autocatalysis.

upmfit has: function upmbuilder() to create the JAGS model script and display a (possibly redefined) design matrix from a design matrix input by the user; upmrun() to generate posterior parameter estimates via a Markov Chain Monte Carlo sampler; and synthetic dataset upmdata.

For a longer introduction, see the introductory vignette for this package. Use command vignette("upm-primer", package="upmfit") or browseVignettes(package = "upmfit") to access the vignettes.

upmfit documentation built on May 2, 2019, 7:25 a.m.