Announcement

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

  • Year fixed effects in a dynamic fixed effects model

    Hi there,
    I have a balanced panel of US states from 1977-2015 to estimate the impact of state income tax changes on economic growth. I am using the user-written command xtpmg to estimate a dynamic fixed effects model and to compare the results from this approach to those estimated previously in the literature, I would like to add yearly fixed effects to the dynamic fixed effects but I am unable to add factor variables (i.year) and when I enter yearly FEs manually, they are all constrained to zero.

    More intuitively, is there a reason yearly FE's cannot be used here that I'm missing? I understand that the first differences will remove any time invariant state FE's and at first thought the first differences remove the yearly FE in this way as well but it seems to me that while the first differences will help remove persistence over time affecting everyone from some shock, they won't capture the effect of the shock on everyone for that year specifically... at least I don't think?

    Thanks so much for your help,
    Bruisah

  • #2
    Big: Statalist mixes nations and cultures and can't be omniscient about naming habits but nevertheless

    http://www.statalist.org/forums/help#realnames

    http://www.statalist.org/forums/help#adviceextras #3

    explains a policy that may apply to you.

    Comment


    • #3
      You're using a user-written routine. Several Stata provided routlines handle dynamic models and may have better documentation/support like xtabond and related routines. There is also user-written xtdpdml.

      Comment

      Working...
      X