Can we use the did_multiplegt package on firm level data in Stata?

[Laurence Jones]

I have firm level panel data, spread over multiple countries. The countries are subject to reforms (aka treatments) occurring at different times (similar to Brodeur et al. (2021)).

Is it possible to use the did_multiplegt stata package to analyse the impact of country level treatments on firm level panel data?

In the documentation it states:

  did_multiplegt Y G T D
Y is the outcome variable.

G is the group variable.

T is the time period variable. The command assumes that the time variable is evenly spaced (e.g.: the panel is at the yearly level, and no year is missing for all groups). When it is not (e.g.: the panel is at the yearly level, but t secutive years are missing for all groups), the command may still be used, though it may require a bit of tweaking. In such instances, contacting the command's authors may be advisable.

D is the treatment variable.

But the issue is that my G is firm level data, but my treatment is on a country level, is it still possible to apply this approach?

When I run a typical DiD type regression my data has 78k observations, but the output from did_multiplegt states that N=456, where are my observations going?

Many thanks

[Jared Greathouse]

Of course. Now, the treated units are based in counties after a given time, else 0. If all 2 police departments in city X are treated after 2019, then let's just compare these two to the other 10 departments in cities x y and z that weren't treated.

[Jared Greathouse]

You'll need to read the documentation on the command itself. Specifically the AER paper.