Hello,
Thanks a lot for your answer.
I have a question about fixed effect
I want to do a multivariate regression and I want to incorporate fixed effects by year and by company here is my equation:
πΆπ΄π =π½ +π½ππ΄ππΆ+π½ππππΈπ +π½πΌππ·+π½π΄πππΌπ·+π½ππΈπ π+π½πΌππ·ππΏπΊ
- πππππ‘πππ(i,t) + Ξ±i + Ξ΄t + Ξ΅i,t
Ξ±i = fixed effect by firms
Ξ΄t = fixed effect by years
Ξ΅i,t = error
control = control variables
My question is the following: I know that there must be only unique peers in my database. But, unfortunately, I sometimes have the same two companies for two similar years (it is not a duplicate, it may be that the first one concerns the beginning of the year and the other one the middle of the year, but we are still in the same year). Can I still use the fixed effect?
Just another information.
I have 2752 unique firms in a sample of 7183 observations.
Is my approach the right one?
I must admit that I am a novice in this field.
Do you have any suggestions?
Thanks you in advance