Dear Statalist users,

I need to
prove that the treatment group and the control group follow similar trends
(parallel trend assumption) prior to the treatment (attending a concert in this case). Someone suggested that I should run
a regression with so-called leads and lags. If the trend is parallel, the leads should not be statistically different from zero. Do you know how to set up the regression with leads and lags in my case? Any help would be greatly appreciated!

I am currently working on a difference in differences estimation using the Taxi and Limousine dataset of 2013. I am trying to estimate if there is impact on the tipping percentage received by taxi drivers if people attend a classical music concert (NewYork Philharmonic).
Therefore, I have created 2 dummy variables:
The first dummy variable is a before/after variable "after_concert" ( within 30min before concert = 0, and within 30min after concert = 1) and the second dummy variable is the geographical location "near_concert" ( anything between >100 meter & <=1000 meter from the entrance of the concert hall = 0, within 100 meter radius from the entrance of the concert hall = 1, this is based on precise GPS coordinates) and an interaction variable "interaction" which is "after_concert" * "near_concert".


There are 3 groups in this study, which are on 3 days in which there is a control and treatment group:

- on the 03jan2013, the "after_concert" dummy is
=0 when the time ranges from 19:00 to 19:30
= 1 when the time ranges from 21:30 to 22.00. There is a 2 hour time gap (19:30 to 21:30) in which there is no data because the concert/treatment happens in that time slot.
- on the 16jan2013, same as on the 03jan2013
- on the 24jan2013, same as on the 03jan2013



Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input float fare_amount double datetime_num float(date time trip_distance after true interaction tip_percentage)
 6.5 1.6728606e+12 19361      19.5 1.24 0 0 0        0
   6 1.6728606e+12 19361      19.5   .9 0 0 0        0
 3.5 1.6728678e+12 19361      21.5  .39 1 1 1 26.66667
   8 1.6728678e+12 19361      21.5 1.76 1 1 1        0
 4.5 1.6728678e+12 19361      21.5  .47 1 0 0        0
 3.5 1.6728678e+12 19361      21.5  .43 1 1 1        0
10.5 1.6728678e+12 19361      21.5 2.53 1 0 0 8.695652
 5.5 1672867840000 19361  21.51111  1.1 1 0 0       20
14.5 1672867860000 19361 21.516666 4.02 1 1 1        0
 5.5 1672867881000 19361   21.5225   .9 1 1 1        0
11.5 1672867902000 19361 21.528334    3 1 1 1       20
11.5 1672867917000 19361   21.5325  2.8 1 1 1        0
end
format %tc datetime_num
format %td date