sparsemediation

The goal of sparsemediation is to conduct sparse mediation analysis.

Installation

You can install the released version of sparsemediation from CRAN with:

install.packages("sparsemediation")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("muschellij2/sparsemediation")

Usage

The sparse.mediation function conducts sparse mediation for the specified tuning parameters for elastic net. Cross-validation is possible using cv.sparse.mediation.

library(sparsemediation)
N = 100
V = 50
set.seed(1234)
a = rbinom(V, 1, 0.1) * 5
b <- a
X = scale(rnorm(N))
M =  X %*% t(a) + matrix(rnorm(N * V), N, V)
Y =  10 * X + M %*% b + rnorm(N)

cvfit <- cv.sparse.mediation(
    X,
    M,
    Y,
    tol = 10 ^ (-10),
    K = 4,
    max.iter = 100,
    lambda = log(1 + (1:10) / 25),
    tau = c(0.5, 1, 2),
    multicore = 4,
    seednum = 1e+06
  )

# fit <- sparse.mediation(
#     X,
#     M,
#     Y,
#     tol = 10 ^ (-10),
#     max.iter = 100,
#     lambda = cvfit$cv.lambda,
#     tau = cvfit$cv.tau
#   )