Bayesian Analysis of Heterogeneous Mediation (BAHM)

This project hosts code for Bayesian heterogeneous mediation described in

Tatiana L Dyachenko, Greg M Allenby (2023), Is Your Sample Truly Mediating? Bayesian Analysis of Heterogeneous Mediation (BAHM), Journal of Consumer Research, 50(1), p.116-141. https://doi.org/10.1093/jcr/ucac041

(new code for categorical outcome (y) and mediator (m) variables is coming soon)

You can get an R package "bahm" following the instructions below. This will open the application on your device.

1. You need to have R installed on your computer:

   • go to https://cran.r-project.org/mirrors.html
   • select a mirror/location and download and install R software
   • you can also install R and RStudio (which provide some convenience) from one place: https://posit.co/download/rstudio-desktop/

2. Open file called LinesToLoad_BAHM.R and follow the instructions in the file (3 lines)

## Install needed libraries 
install.packages(c("shiny", "bayesm", "HDInterval", "coda", "gtools","dplyr", "readr", "DT", "future", "promises", "ipc", "future.callr",
                   "future.apply", "data.table", "shinyWidgets", "devtools","htmltools", "shinydashboard","shinyjs"))

## Getting the package from github
devtools::install_github("tdyachenko/BAHM-Mediation", ref = "r-pkg",force = TRUE)

## Loading the package 
library(bahm)

## Starting the app on your device
run_bahm()

Alternatively, the app can be run as a web application, but it might be much slower. https://bayesianmediationanalysis.shinyapps.io/BAHM/