https://github.com/carloscinelli/dml.sensemakr
Sensitivity analysis tools for causal ML
https://github.com/carloscinelli/dml.sensemakr
causal-inference debiased-machine-learning double-machine-learning machine-learning r r-package sensitivity-analysis
Last synced: 5 months ago
JSON representation
Sensitivity analysis tools for causal ML
- Host: GitHub
- URL: https://github.com/carloscinelli/dml.sensemakr
- Owner: carloscinelli
- Created: 2022-09-06T02:02:45.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2025-07-25T07:55:50.000Z (10 months ago)
- Last Synced: 2025-07-25T13:30:36.037Z (10 months ago)
- Topics: causal-inference, debiased-machine-learning, double-machine-learning, machine-learning, r, r-package, sensitivity-analysis
- Language: R
- Homepage: http://carloscinelli.com/dml.sensemakr/
- Size: 145 MB
- Stars: 17
- Watchers: 2
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.Rmd
Awesome Lists containing this project
README
---
output: github_document
---
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
cache = F,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```
# dml.sensemakr
`dml.sensemakr` implements a general suite of sensitivity analysis tools for Causal Machine Learning as discussed in [Chernozhukov, V., Cinelli, C., Newey, W., Sharma A., and Syrgkanis, V. (2021). "Long Story Short: Omitted Variable Bias in Causal Machine Learning."](https://www.nber.org/papers/w30302)
# Development version
To install the development version on GitHub make sure you have the package devtools installed.
```{r, eval = FALSE}
# install.packages("devtools")
devtools::install_github("carloscinelli/dml.sensemakr")
```
# CRAN
CRAN version coming soon.
# Details
For theoretical details [please see the working paper.](https://www.nber.org/papers/w30302)
For a primer on Debiased Machine Learning,
[please check Chernozukohv et al (2018).](https://academic.oup.com/ectj/article/21/1/C1/5056401)
Some presentations that may be useful:
- [Carlos' presentation at the ICLR.](http://interactivecausallearning.com/Carlos_Cinelli.html)
- [Victor's tutorial at the Chamberlain Seminar.](https://www.youtube.com/watch?v=PQtYqKfxH_I)
# Basic Usage
```{r basic-usage, fig.align='center', collapse=T, dpi=300}
# loads package
library(dml.sensemakr)
## loads data
data("pension")
# set treatment, outcome and covariates
y <- pension$net_tfa # net total financial assets
d <- pension$e401 # 401K eligibility
x <- model.matrix(~ -1 + age + inc + educ+ fsize + marr + twoearn + pira + hown, data = pension)
# run DML (nonparametric model)
dml.401k <- dml(y, d, x, model = "npm")
# summary of results with median method (default)
summary(dml.401k)
# sensitivity analysis
sens.401k <- sensemakr(dml.401k, cf.y = 0.04, cf.d = 0.03)
# summary
summary(sens.401k)
# contout plots
plot(sens.401k)
```