Hello,
I have read all the DiD posts and help forums and still can't figure out the answer to my question.
I have a sample of firms that have received one treatment in different years between 1990-2010 (i.e., staggered treatments) and I am not matching my sample (i.e., creating a typical control group as most DiD experiments do). Thus, my entire sample consists of firms that were treated at one point in time during the 1990-2010 window, with some firms being treated very early (e.g., 1993) and some quite late (e.g., 2007). I have data for every firm in every year.
I want to test the effect of the treatment on the treated firms over a 4 year observation window (change from 2 years before the treatment to 2 years after). Can the "control" group consist of all firms that have not yet received the treatment? That is, a firm is in the control group until it is treated, at which time it joins the treatment group.
In the past I have matched the treated firms with control firms but I am not sure that is necessary here since I have a large sample of firms to which I can compare the treated firms to (the firms that had not yet been treated in that year).
Thank you in advance for your help! I tried to be as clear as possible and can add more detail if necessary.
Sincerely,
Roger
I have read all the DiD posts and help forums and still can't figure out the answer to my question.
I have a sample of firms that have received one treatment in different years between 1990-2010 (i.e., staggered treatments) and I am not matching my sample (i.e., creating a typical control group as most DiD experiments do). Thus, my entire sample consists of firms that were treated at one point in time during the 1990-2010 window, with some firms being treated very early (e.g., 1993) and some quite late (e.g., 2007). I have data for every firm in every year.
I want to test the effect of the treatment on the treated firms over a 4 year observation window (change from 2 years before the treatment to 2 years after). Can the "control" group consist of all firms that have not yet received the treatment? That is, a firm is in the control group until it is treated, at which time it joins the treatment group.
In the past I have matched the treated firms with control firms but I am not sure that is necessary here since I have a large sample of firms to which I can compare the treated firms to (the firms that had not yet been treated in that year).
Thank you in advance for your help! I tried to be as clear as possible and can add more detail if necessary.
Sincerely,
Roger