SCPI

The scpi package provides Python, R and Stata implementations of estimation and inference procedures for synthetic control methods.

This work was supported by the National Science Foundation through grants SES-1947805 and SES-2019432, and by the National Institutes of Health through grant R01 GM072611-16.

Queries and Requests

Please email: scpi_pkg@googlegroups.com

Python Implementation

To install/update in Python type:

pip install scpi_pkg

• Help: PyPI repository.

• Replication: py-script, plot illustration, data.

• Illustration Staggered Adoption: py-script, plot illustration.

R Implementation

To install/update in R from CRAN type:

install.packages('scpi')

• Help: R Manual, CRAN repository.

• Replication: R-script, plot illustration, data.

• Illustration Staggered Adoption: R-script, plot illustration.

Stata Implementation

The Stata implementation relies on Python, which needs to be available in the system.

How to install Python

There are at least two ways to install Python:

1. Download and install Python directly from https://realpython.com/installing-python/.
2. Download and install Anaconda for Windows, macOS, or Linux.

After Python is installed, please install the scpi package in Python using the instructions above.

How to link Stata and Python

Stata (16.0 or newer) and Python (>=3.8) can be linked following the official tutorial on the Stata blog.

To install/update in Stata type:

net install scpi, from(https://raw.githubusercontent.com/nppackages/scpi/master/stata) replace

• Help: scdata, scest, scpi, scplot.

• Replication files: do-file, plot illustration, data.

• Illustration Staggered Adoption: do-file, plot illustration.

References

Software and Implementation

• Cattaneo, Feng, Palomba and Titiunik (2022): scpi: Uncertainty Quantification for Synthetic Control Estimators.
  Working paper.

Technical and Methodological

• Cattaneo, Feng, Palomba and Titiunik (2022): Uncertainty Quantification in Synthetic Controls with Staggered Treatment Adoption.
  Working paper (coming soon).

• Cattaneo, Feng and Titiunik (2021): Prediction Intervals for Synthetic Control Methods.
  Journal of the American Statistical Association 116(536): 1865-1880.
  Supplemental