reycn/FSynth

An R package offering speed enhancements for Synthetic Control.

Faster Synthetic Control

An R package offering speed enhancements for Synthetic Control.

Authors: Rongxin Ouyang, Subhayan Mukerjee
Last Update: Oct., 2024

Performance

• Normalizer: time lapsed - 98.67%, consistently improving with larger datasets
• Full process: -8%

Usage

1. Clone this repository: git clone https://github.com/reycn/FSynth.git

2. In R / notebook, load locally

   # Install devtools to load local packages
   if (!require("pacman")) install.packages("pacman")
   pacman::p_load(devtools)
   # Load FSynth from a local path
   devtools::load_all("/local/path/to/repo")

3. Use this package in the same way to the original package, Synth (Abadie, 2010)

   dataprep()
   synth()

Changes

• We vectorized the normalization process
• We used .Primitive("func") functions

Our Paper

• Link [TBD]

Citation

• [TBD]

Reference

• Abadie, A., Diamond, A., & Hainmueller, J. (2011). Synth: An R package for synthetic control methods in comparative case studies. Journal of Statistical Software, 42(13).