Nonparametric Bounds for the Average Causal Effect Due to Balke and Pearl and Extensions

Implementation of the nonparametric bounds for the average causal effect under an instrumental variable model by Balke and Pearl (Bounds on Treatment Effects from Studies with Imperfect Compliance, JASA, 1997, 92, 439, 1171-1176, ). The package can calculate bounds for a binary outcome, a binary treatment/phenotype, and an instrument with either 2 or 3 categories. The package implements bounds for situations where these 3 variables are measured in the same dataset (trivariate data) or where the outcome and instrument are measured in one study and the treatment/phenotype and instrument are measured in another study (bivariate data).

bpbounds: Nonparametric bounds for the average causal effect for one-sample (bivariate) and two-sample (trivariate) data

R package implementing the nonparametric bounds for the average causal effect of Balke and Pearl (1997).

Installation

Install the released version of bpbounds from CRAN:

install.packages("bpbounds")

or from the MRCIEU R-universe

install.packages("bpbounds", repos = c("https://mrcieu.r-universe.dev", "https://cloud.r-project.org"))

Or install the development version from my r-universe

install.packages("bpbounds", repos = c("https://remlapmot.r-universe.dev", "https://cloud.r-project.org"))

Or install the development version from GitHub with:

# install.packages("remotes") 
remotes::install_github("remlapmot/bpbounds")

Shiny App

There is a Shiny app demonstrating the package at: https://remlapmot.shinyapps.io/bpbounds/.

Package website

The helpfiles and vignette are shown on the package website at: https://remlapmot.github.io/bpbounds/.

Authors

Tom Palmer (maintainer, [email protected]), Roland Ramsahai, Vanessa Didelez, Nuala Sheehan.

References

Balke A, Pearl J. Bounds on Treatment Effects from studies with imperfect compliance. Journal of the American Statistical Association, 1997, 92, 439, 1171-1176, doi: 10.1080/01621459.1997.10474074.