Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • diff in diff, unbalanced panel data re election years

    Hello! I am trying to do a diff in diff with fixed effects on unbalanced panel data. My dependent variable is individual contributions to state-level campaigns w my time fixed effect being whether or not the year is during a resource boom and entity fixed effect equaling state. Attached is the tabulation of year/election with election=1 and special election=2. A handful of states have off-year elections and it seems to be affecting my data. For the regression attached I used reg contr_rate top10 i.year if(elect==1), r
    Am I approaching this correctly? What is the best way to deal with the off-year election? Thank you
    Attached Files

  • #2
    I don't understand your explanation of what you want to do. That said, the model you have used in your regression contains no interaction terms, so it is not a diff-in-diff model (unless the variable top_ten is some kind of disguised interaction--but in that case the components of the interaction are missing and the model is mis-specified.) So while I am pretty sure what you have done is wrong, without a clearer understanding of your data and your goals, I can't advise how to fix it.

    Comment


    • #3
      I am not sure if I understand what your are going for either but you may try looking at this explanation for doing diff-in-diff in Stata. If you can write out a list of your commands, it can be easier to advise.

      http://www.princeton.edu/~otorres/DID101.pdf

      Comment

      Working...
      X