I am running a two way fixed effects regression with state and year fixed effects. I would like to add a state-specific linear time trend and so I have a few questions on this:
1. I believe this can be achieved with interacting state with year in addition to having the individual state and year fixed effects. What is the correct way to specify this?
Code:
reghdfe outcome treatment, absorb(c.year#i.statefip) vce(cluster statefip)
Code:
reghdfe outcome treatment, absorb(i.year#i.statefip) vce(cluster statefip)
2. When specified as i.year#i.statefip, the treatment variable is dropped because of collinearity. Would you happen to know why this is happening and if so, is there a way around it? Is it because i.year is stripping out a lot of variation?
3. Finally, how would I go about adding a quadratic state year trend in light of the above?
Many thanks in advance for your help!
0 Response to Specifying state-specific linear time trends and potential issues of collinearity
Post a Comment