Dear all,
I am trying to estimate the difference of impact of the Athenes Olympic and the Paralympic games in on tourism for my bachelor thesis. Usually, previous litterature use the gravity model but it´s not possible in this case as I cannot find the monthly data needed. Anyway, I was thinking of running a difference-in-difference, in this case comparing monthly tourist arrivals between hosting [treatment group] and non hosting countries [control group, including spain, italy and croatia]. I would like to have the difference before the event, during the event and after the event. It is shown in my data file by the column timeOG, where timeOG=0 for the period before the Olympics, =1 during the Olympics in august and =2 after the Olympics.
But when I do the diff-in-diff on stata I have several problems :
- First I only have two periods, before and after. Stata considers that it is a binary variable. Is there any way that it could be turned into a "triple variable" ?
- And I would like to include a fixed effect for country but I don´t know if it can be done in the case of a diff-in-diff nor have to do it.
Here are snapshot of my stata file. I am quite new in econometrics so these questions might seem futile but it would help me a lot. Don´t hesitate to tell me if it´s not the right method or anything, I can´t tell by myself.
Thank you very much.
I am trying to estimate the difference of impact of the Athenes Olympic and the Paralympic games in on tourism for my bachelor thesis. Usually, previous litterature use the gravity model but it´s not possible in this case as I cannot find the monthly data needed. Anyway, I was thinking of running a difference-in-difference, in this case comparing monthly tourist arrivals between hosting [treatment group] and non hosting countries [control group, including spain, italy and croatia]. I would like to have the difference before the event, during the event and after the event. It is shown in my data file by the column timeOG, where timeOG=0 for the period before the Olympics, =1 during the Olympics in august and =2 after the Olympics.
But when I do the diff-in-diff on stata I have several problems :
- First I only have two periods, before and after. Stata considers that it is a binary variable. Is there any way that it could be turned into a "triple variable" ?
- And I would like to include a fixed effect for country but I don´t know if it can be done in the case of a diff-in-diff nor have to do it.
Here are snapshot of my stata file. I am quite new in econometrics so these questions might seem futile but it would help me a lot. Don´t hesitate to tell me if it´s not the right method or anything, I can´t tell by myself.
Thank you very much.
