missingHE: Missing Cost Effectiveness Models

Fit Bayesian Missing Cost Effectiveness Models in R.

:rocket: This is the development version of the R package missingHE (currently on version 1.6.3). The stable version is now release 1.6.2, on CRAN.

Contents

• Overview
• Features
• Installation
• Further details

Overview

Given a individual-level partially-observed cost-effectiveness outcome data in the form of a dataframe, fits different types of Bayesian models to handle missing effect and/or cost values under different assumptions about the missing data mechanisms, assesses model performance and Markov chain Monte Carlo (MCMC) convergence, and summarises statistical and economic results based on the posterior distribution of key model parameters.

Features

Main features of missingHE include:

• Analysis of partially-observed effects/costs data via Bayesian missingness models
• Model assessment measures/plots and MCMC convergence checks
• Posterior summaries for key model parameters and cost-effectiveness measures
• Standard cost-effectiveness measures/graphs based on model results

Installation

Install the released version from CRAN with

install.packages("missingHE")

The development version (in this repo, which can be updated more quickly and more often than the stable one) can be installed using the following command

pkgs <- c("R2jags", "ggplot2", "gridExtra", "BCEA", "ggmcmc", "loo", "Rtools", "devtools", "utils")
repos <- c("https://cran.rstudio.com") 
install.packages(pkgs, repos = repos, dependencies = "Depends")

before installing the package using devtools:

devtools::install_github("AnGabrio/missingHE", build_vignettes = TRUE)

The optional argument build_vignettes = TRUE allows to install the vignettes of the package locally on your computer. These consist in brief tutorials to guide the user on how to use and customise the models in missingHE using different functions of the package. Once the package is installed, they can be accessed using the command

utils::browseVignettes(package = "missingHE")

which shows all the vignettes available for the package. Alternatively, you can install the development version using remotes, with the following command.

install.packages("remotes")
remotes::install_github("AnGabrio/missingHE")

NB: On Windows machines, you need to install a few dependencies, including Rtools first, e.g. by running

pkgs <- c("MASS", "Rtools", "remotes")
repos <- "https://cran.rstudio.com"
install.packages(pkgs, repos = repos, dependencies = "Depends")

before installing the package using remotes.

All models implemented in missingHE are written in the BUGS language using the software JAGS, which needs to be installed from its own repository and instructions for installations under different OS can be found online. Once installed, the software is called in missingHE via the R package R2jags. Note that the missingHE package is currently under active development and therefore it is advisable to re-install the package directly from GitHub before each use to ensure that you are using the most updated version.

Further details

The pkgdown site is here. Also, details about the package, including some references and links to a pdf presentation and some posts on my own blog are given here.

License

Contributing

Please submit contributions through Pull Requests, following the contributing guidelines. To report issues and/or seek support, please file a new ticket in the issue tracker.

Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.