I've to add industry effect and year effect in fixed effect model where my panels are based on firms. When I add industry dummy, results omit the industry dummies because of collinearity.
Can I use an industry*year effect together as I saw it in some studies? If yes, how to incorporate it in model? I tried it in two ways; first by creating an interaction term of both industry and year and then using it as i.industryXyear variable in the model and I got the error saying "matsize too small".
Second, I tried this code for interaction effect (note:I applied it for 12 industries and 41 years)
Code:
Code:
xi i.ffind12*fiscalyear
i.ffind12 _Iffind12_1-12 (naturally coded; _Iffind12_1 omitted)
i.ffi~12*fis~ar _IffiXfisc_# (coded as above)
and I ended up in 12 dummies. I don't know the validity of code and how I can incorporate it in fixed effect model?
Would be grateful for any help.
0 Response to Industry effect and year effect together in fixed effect model
Post a Comment