Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Estimation of the difference-in-difference model with multiple time periods, using a series of months

    Hello everyone,

    I am carrying out an investigation into how the signing of the Multiparty Trade Agreement between the EU and the Andean countries affected their exposures. For that I am using the Callaway and Santana difference-in-difference model. The countries with which I am working are Ecuador, Colombia and Peru, each of them signed the agreement in different periods of time. This is how the first to sign was Peru in March 2013, then Colombia in August 2013 and finally Ecuador in January 2017. Therefore, the counterfactual in this case is those that have not yet been treated. Therefore, the counterfactual for Peru would be Colombia and Ecuador, and for Colombia it would be Ecuador. The problem that arises for me is that when I make the estimate using the following command, I get this table and I don't know if it is correct or when I have an estimate by month and year I should put something else in the command.

    Code:
     
    csdid VolPeso_TM, ivar(Codpais) time(Fecha2) gvar(firsttreat3) method(dripw) notyet
    Where the variable "VolPeso_TM" is the volume of exports. The variable "Codpais" is the code for each country 1 ECUADOR, 2 COLOMBIA AND 3 PERU. The variable "Date3" is the numerical variable that goes from 1 to 119, 1 represents January 2013 and 119, which is the end, represents November 2022. That is, this variable collects the date from January 2013 to November 2022. Finally, the variable " firsttreat3" is the variable that takes the treatment.


    Here I attach the image of the estimate and a base with which I am making the estimate. exp.dta

    Click image for larger version

Name:	Captura de pantalla 2024-02-27 170702.png
Views:	1
Size:	38.2 KB
ID:	1744830
Working...
X