-xtevent- Stata module to estimate and visualize linear panel event-study models now available on SSC

[Jorge Eduardo Perez Perez]

It's been a while since I posted anything to Statalist!!!

Thanks to Kit Baum, -xtevent- is now available on SSC.

-xtevent- provides commands to estimate linear panel event-study models, create event-study plots, and conduct hypothesis tests after estimation. It implements the plotting suggestions and many of the estimators discussed in Freyaldenhoven et al. (2021).

Here's the companion paper:

https://www.nber.org/papers/w29170

Here's an ungated version:

https://www.brown.edu/Research/Shapi...EventStudy.pdf

Here's a Twitter thread summarizing the paper in 4 short tweets:

https://twitter.com/jorpppp/status/1430592541083062272

And here you can see it in action!

[John Singer]

Hi Jorge Eduardo Perez Perez , thank you for creating this! I am curious to learn more about the policyvar variable. If I have a policy that went into effect in a given year but I don't have any other specific information data (other than year), can I still use this module? Do I make a dummy for the year? Or is it not possible to use the module in this case? And how should I set up the window variable? I keep on getting the error "Window outside event-time range" r(198), no matter what number I put into the window().

[Iuri Gavronski]

It's been a while since I posted anything to Statalist!!!

Thanks to Kit Baum, -xtevent- is now available on SSC.

-xtevent- provides commands to estimate linear panel event-study models, create event-study plots, and conduct hypothesis tests after estimation. It implements the plotting suggestions and many of the estimators discussed in Freyaldenhoven et al. (2021).

Here's the companion paper:

https://www.nber.org/papers/w29170

Here's an ungated version:

https://www.brown.edu/Research/Shapi...EventStudy.pdf

Here's a Twitter thread summarizing the paper in 4 short tweets:

https://twitter.com/jorpppp/status/1430592541083062272

And here you can see it in action!
[ATTACH=CONFIG]n1627509[/ATTACH]

Jorge Eduardo Perez Perez sorry for the question, I am not familiarized with event studies. -xtevent- expects a dichotomous variable for the policy. Can you point to a good reference on how to "convert" a continuous variable to an "event", say, a change in direction in R&D investments or something similar?

[fabian scheifele]

Dear Jorge Eduardo Perez Perez, I have a question regarding the window option: I have a balanced panel dataset from 2004 to 2020 (and 113 firms as panel variable) and I have also created my policy dummy variable that indicates the firm-year instance where the event took place, similar to the union variable in the example code. Hence, I have 17 time periods in total and the event starts in 2011 and last til 2020 so 10 time periods. When I put w(8) for symmetric time windows or w (-7 9) for assymetric, I get the error "Window outside event-time range". Similarly I also tried to use the year integers to identify the window (i.e. w(2004 2020)) but i get the error "Window can not be negative". I checked and everything else is as in the example (xtset specified, policy variable specified). Do you have an idea what could be wrong with my window specification?

[Rafiuddin Najam]

Dear Jorge Eduardo Perez Perez, I have a question regarding the window option: I have a balanced panel dataset from 2004 to 2020 (and 113 firms as panel variable) and I have also created my policy dummy variable that indicates the firm-year instance where the event took place, similar to the union variable in the example code. Hence, I have 17 time periods in total and the event starts in 2011 and last til 2020 so 10 time periods. When I put w(8) for symmetric time windows or w (-7 9) for assymetric, I get the error "Window outside event-time range". Similarly I also tried to use the year integers to identify the window (i.e. w(2004 2020)) but i get the error "Window can not be negative". I checked and everything else is as in the example (xtset specified, policy variable specified). Do you have an idea what could be wrong with my window specification?

I had the same issue. You may want to try window(-6 8)

[Jorge Eduardo Perez Perez]

Jorge Eduardo Perez Perez sorry for the question, I am not familiarized with event studies. -xtevent- expects a dichotomous variable for the policy. Can you point to a good reference on how to "convert" a continuous variable to an "event", say, a change in direction in R&D investments or something similar?

Iuri Gavronski -xtevent- works with continuous policy variables too!

[Jorge Eduardo Perez Perez]

Hi Jorge Eduardo Perez Perez , thank you for creating this! I am curious to learn more about the policyvar variable. If I have a policy that went into effect in a given year but I don't have any other specific information data (other than year), can I still use this module? Do I make a dummy for the year? Or is it not possible to use the module in this case? And how should I set up the window variable? I keep on getting the error "Window outside event-time range" r(198), no matter what number I put into the window().

John Singer The policy variable in this case should be an interaction of a dummy for time after the policy and a dummy for the treated units.

[Jorge Eduardo Perez Perez]

Thanks to everyone who has been following this thread. Check out the thread for the new version of -xtevent- here:

https://www.statalist.org/forums/for...updated-on-ssc