Generates different posterior distributions of adjusted odds ratio under different priors of sensitivity and specificity, and plots the models for comparison. It also provides estimations for the specifications of the models using diagnostics of exposure status with a non-linear mixed effects model. It implements the methods that are first proposed in
R package for accounting for misclassification with Bayesian methods Author: James (Jinhui) Yang, Haitao Chu, Lifeng Lin
This package is based on the methods proposed in
BayesSenMC provides users with 6 different models (as outlined in Chu's paper) that estimates the adjusted odds ratio in a case-control study based on the different priors of sensitivity and specificity as well as historic data of related studies.
The models with zero and constant misclassification (constant Se and Sp) are able to be used given the 2x2 table of a case-control study. However, other models require parameters from a NLMIXED procedure. Similarly, the graphing function can be used with stanfit objects that are returned from any of the model functions.