This is an implementation of a more 'flexible' version of the synthetic controls method for case studies and causal impact evaluation.
The traditional synthetic controls implemented by Abadie, Diamond, and Hainmueller (2010) has 5 constraints placed on the output weights:
However, it is not necessary in all cases to hold all of these constraints.
This method, published by Doudchenko and Imbens in 2016, is an implementation of Synthetic Controls that only holds two of these constraints:
by using glmnet as the underlying optimization.
Install with the following command
devtools::install_github('tdn158/ElasticSynth')
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.