The goal of ccmEstimator is to perform comparative causal mediation analysis to compare the mediation effects of different treatments via a common mediator.
You can install the released version of ccmEstimator from CRAN with:
install.packages("ccmEstimator")This is a basic example which shows you how to solve a common problem:
library(ccmEstimator)
## basic example code
# load example data. ICAapp is a data frame with 14 variables. It is the application data used to illustrate comparative causal mediation analysis methods in Bansak (2020).
data(ICAapp)
set.seed(321, kind = "Mersenne-Twister", normal.kind = "Inversion")
ccm.results <-
getCCM(Y = "dapprp", T1 = "trt1", T2 = "trt2", M = "immorp", data = ICAapp,
noInteraction = TRUE, sigLevel = 0.05, boots = 1000)
#> Estimation assuming no interactions between treatments and mediator.
#> The estimated mediation effect for treatment 2 is 1.563 times larger than the mediation effect for treatment 1
#> (with 95% CI: [1.183,2.16])
#> The estimated proportion mediated for treatment 2 is 0.952 times the size of that for treatment 1
#> (with 95% CI: [0.755,1.236])
#> Please use summary() to view the results in more detail.
summary(ccm.results)
#> Call:
#> getCCM(Y = "dapprp", T1 = "trt1", T2 = "trt2", M = "immorp",
#> data = ICAapp, noInteraction = TRUE, sigLevel = 0.05, boots = 1000)
#>
#> point.estimate lower.ci upper.ci
#> ATE1 0.1949336 0.1444899 0.2509022
#> ATE2 0.3201672 0.2655617 0.3794662
#> ACME1 0.1132405 0.07559316 0.1502631
#> ACME2 0.1769843 0.1382323 0.2186979
#>
#> point.estimate lower.ci upper.ci
#> ACME2/ACME1 1.562907 1.182749 2.159821
#> (ACME2/ATE2)/(ACME1/ATE1) 0.9515747 0.7549031 1.235673