hdbm is a Bayesian inference method that uses continuous shrinkage priors for high-dimensional mediation analysis, developed by Song et al (2018). hdbm provides estimates for the regression coefficients as well as the posterior inclusion probability for ranking mediators.


You can install hdbm via CRAN


Or devtools

devtools::install_github("umich-cphds/hdbm", build_opts = c())

If you wish to install the package via devtools, you will need a C++ compiler installed. This can be accomplished by installing Rtools on Windows and Xcode on MacOS.


Taken from the hdbm help file


Y <-$y
A <-$a

# grab the mediators from the example data.frame
M <- as.matrix([, paste0("m", 1:100)], nrow(

# We just include the intercept term in this example.
C <- matrix(1, nrow(M), 1)
beta.m <- rep(0, 100)
alpha.a <- rep(0, 100)

output <- hdbm(Y, A, M, C, C, beta.m, alpha.a, burnin = 3000, ndraws = 100)

# Which mediators are active?
active <- which(colSums(output$r1 * output$r3) > 50)


Yanyi Song, Xiang Zhou et al. Bayesian Shrinkage Estimation of High Dimensional Causal Mediation Effects in Omics Studies. bioRxiv 467399

Try the hdbm package in your browser

Any scripts or data that you put into this service are public.

hdbm documentation built on Aug. 28, 2019, 5:04 p.m.