Need help with how to run Synthetic Control Method for Staggred DID treatment

[Saleh sharmah]

Hi users

I am new to using the Synthetic control method as outlined in Ben-Michael et al. [2021]. . I am interested in examining the impact of a mandate on a firm's transparency. The mandate is staggered across different countries and I would like to use this method to plot the annual estimates of the average treatment effect on the treated, including 95% confidence intervals, in event-time relative to the year of adoption of the disclosure regulation: beow is the an example of my data:

input float(transparency treatedxmandate) int(Policy1Year Year) float firm_id
.97 0 . 2010 7
1.01 0 . 2013 7
.39 0 . 2014 1561
.98 0 . 2014 4
.99 0 . 2014 7
.99 0 . 2015 7
.98 0 . 2015 4
.42 0 . 2015 1561
.51 0 . 2015 6
.88 0 . 2015 5
.98 0 . 2016 4
.53 0 . 2016 5
.45 0 . 2016 7
.32 0 . 2016 6
.35 0 . 2016 1561
.53999996 0 . 2017 5
.99 0 . 2017 4
.31 0 . 2017 6
.4 0 . 2017 7
1.02 0 . 2018 8
.39 0 . 2018 7
.55999994 0 . 2018 5
.31 0 . 2018 6
.99 0 . 2018 4
.39 0 . 2019 7
.3 0 . 2019 6
.57 0 . 2019 5
.99 0 . 2019 4
1.04 0 . 2019 8
.98 0 . 2020 5
.13 0 . 2005 17
.52 0 . 2005 18
.52 0 . 2005 16
.6 0 . 2006 16
.78 0 . 2006 1722
.36 0 . 2006 17
.53 0 . 2006 18
.34 0 . 2006 22
.26999998 0 . 2007 17
.56 0 . 2007 18
.34 0 . 2007 22
.4 0 . 2008 17
.5 0 . 2008 18

thanks for your kind help.

[Jared Greathouse]

Augsynth is an R command. You'd need to look at their recent paper in order to see how they do it. You can email Eli for the code.

Either way, why are your data not sorted in panel data form? How many donors do you have?

[Saleh sharmah]

Thanks Jared for your kind suggestion.