Hello everyone,
I am using the did_multiplegt command in order to perform some analyses for my thesis and have run into a problem.
I have been running this code:
lan_pass is my outcome variable, cctn identifies my units, year the period, fi_1_e the treatment (a binary treatment) and ci_1_e is a control (binary as well).
I have been trying to get this code to run, but have consistently gotten this error message:
I have been struggling to make sense of this prompt because my treatment is discrete. I have also checked other discussions on this error and found that most relate to lacking a package, but I have installed the ones the help section for the command indicates should be installed already (reghdfe, ftools, moremata). I am also confused because I have already used this data set with other commands (regress, reghdfe, did_imputation).
I include an excerpt of my data so anyone that is familiar with the command can have a look and point out if there is anything about it that is making the command fail:
I am using the did_multiplegt command in order to perform some analyses for my thesis and have run into a problem.
I have been running this code:
Code:
did_multiplegt lan_pass cctn year fi_1_e, controls(ci_1_e).
I have been trying to get this code to run, but have consistently gotten this error message:
The command was not able to estimate the treatment effect at the period when groups' treatment change.
If your treatment is continuous or takes a large number of values, you may need to use
the threshold_stable_treatment option to ensure you have groups whose treatment does not change
over time. You may also need to use the recat_treatment option to discretize your treatment variable.
If your treatment is continuous or takes a large number of values, you may need to use
the threshold_stable_treatment option to ensure you have groups whose treatment does not change
over time. You may also need to use the recat_treatment option to discretize your treatment variable.
I include an excerpt of my data so anyone that is familiar with the command can have a look and point out if there is anything about it that is making the command fail:
Code:
* Example generated by -dataex-. For more info, type help dataex
clear
input float(lan_pass cctn) int year float(fi_1_e ci_1_e)
78.9 1 2009 0 1
. 1 2010 0 1
39.31 1 2011 0 1
. 1 2012 0 1
51.5 1 2013 0 1
. 1 2014 0 1
75 1 2015 0 1
. 1 2016 0 1
60 1 2017 1 1
. 1 2018 1 1
84.6 1 2019 1 1
. 1 2020 1 1
. 1 2021 1 1
. 1 2022 1 1
. 1 . . .
73.73 2 2009 0 0
. 2 2010 0 0
64.49 2 2011 0 0
. 2 2012 0 0
72.3 2 2013 0 0
. 2 2014 0 0
84.3 2 2015 0 1
. 2 2016 0 1
84.3 2 2017 1 1
. 2 2018 1 1
75.6 2 2019 1 1
. 2 2020 1 1
. 2 2021 1 1
. 2 2022 1 1
. 2 . . .
end
Comment