FlagAE: FlagAE: A package for flagging adverse events by Bayesian...
In ganluan123/FlagAE: Flag adverse events by Bayesian methods

Description Details Functions in this package Dataset in this package preinstalled packages required

FlagAE: A package for flagging adverse events by Bayesian methods.

This packages can flag adverse events by three stage Bayesian Hierarchical model (see Hier) and by Bayesian model with Ising latent variable (see Ising). It also provide the option to flag adverase events by frequentist method (see BCIplot). Adverse events selected by both Bayesian methods can be plotted together for comparison (see HIplot). Last this package provides the options to compare these two bayesian models or compare differnt hyperparamters for prior distribution in one model by calculating the loss function with cross validation (see CVhier, CVising).
Two datasets are also provided in this packages for demo purpose. They are ADAE (adverse event analysis dataset), and ADSL (subject level analysis dataset). They can be used for running the example code for each function to give users a better understanding of these functions.

Raw data process: preprocess
Plot based on Binomial confidence interval: BCIplot, BCItable
Three stages Bayesian Hierarchical Model: Hier_history, sum_Hier, Hier, Hiergetpi , Hierplot, Hiertable
Bayesian Model with Ising Latent Variables: Ising_history, sum_Ising, Ising, Isinggetpi , Isingplot, Isingtable
Plot for two Bayesian models: HIplot, HItable
Loss Calculation by Cross Validation: Lossfun, kfdpar, CVhier, CVising