The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
The CERFIT R package is an implementation of Random Forest of Interaction Trees (RFIT) in R. Which is a modification of the random forest algorithm to estimate individualized treatment effect in randomized control trials and observational studies. It does this by modifying the split rule used in the tree growing process. It chooses a split by maximizing subgroup treatment heterogeneity. It can handle binary, multiple, ordered, and continuous treatment.
You can install the development version of CERFIT from GitHub with:
devtools::install_github("JustinThorp/CERFIT")Below is a small example showing how to fit a Random Forest of Interaction Trees, get predictions from the tree, and calculate variable importance. When growing the forest it will print our the response type, the treatment levels, the type of treatment variable, and an updating count of the number of trees.
library(CERFIT)
fit <- CERFIT(Y ~ SAT_MATH + HSGPA + AGE + GENDER + URM | A,
method = "observation",
PropForm = "CBPS",
data = educational,
ntrees = 30)
#> Continous Response
#> Treatment Levels: 1 2 3
#> Multiple Treatment
#> Tree Number: 10
#> Tree Number: 20
#> Tree Number: 30
ite <- predict(fit)
#> [1] 1 2 3
importance <- MinDepth(fit)library(tidyverse)
ite %>%
as.data.frame() %>%
pivot_longer(everything()) %>%
ggplot(aes(x = name, y = value)) +
geom_violin() +
scale_x_discrete(labels = c(0,1,2)) +
labs(y = "Predicted Y",x = "A") +
theme_bw()
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.