Skip to content

This R package calculates various measures of association and helps understand and visualize the link between causal models such as Marginal Structural Models and Standerdized measures.

master
Switch branches/tags
Go to file
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
R
 
 
man
 
 
 
 
 
 
 
 
 
 
 
 

README.md

iptw

R package for calculating Inverse Probability of Treatment Weights

  • This package calculates various measures of association and helps understand and visualize the link between causal models such as Marginal Structural Models and Standerdized measures.

Installation

library(devtools)
install_github("ehsanx/iptw")

Loading the package

require(iptw)

Pulling the help file

help(package="iptw")

Using this package

r1 <- .9 # Risk of the treated subjects
r0 <- .5 # Risk of the untreated subjects
measures.calc(r1,r0)
# Calculates risk measures

form.table.object <- form.table(Y1A1L1=150, 
                                Y1A0L1=45, 
                                Y1A1L0=20, 
                                Y1A0L0=5, 
                                Y0A1L1=300, 
                                Y0A0L1=10, 
                                Y0A1L0=40, 
                                Y0A0L0=55)
form.table.object
# Forms the crude, stratified tables suitable 
# for using in other functions of this package.

ungrouped.data.object <- ungrouped.data(form.table.object)
# Creates ungrouped data suitable for performing regression on it
head(ungrouped.data.object)

associational(form.table.object) 
# Calculates measures of associations
standardization(form.table.object) 
# Calculates Standardized measures
iptw(form.table.object, type = "sw") 
# Standardized measures are calculated by this function 
# using inverse probability of treatment weights (IPTW).

iptw.regression(form.table.object, type = "sw") 
# Calculates IPTW estimates from regression
# requires grid and survey package

graph(form.table.object, type = "sw")
# Maps probabilities and frequesncies

Author

  • Ehsan Karim :octocat: Feel free to report any errors / update suggestions.

Useful Reference

  • Hernán MA, Robins JM (2016). Causal Inference. Boca Raton: Chapman & Hall/CRC, forthcoming.

Related web-Apps

About

This R package calculates various measures of association and helps understand and visualize the link between causal models such as Marginal Structural Models and Standerdized measures.

Resources

License

Packages

No packages published

Languages