Announcement

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

  • FE Pannel regression with controls that don't change over time or firm

    Dear Statalist, I am trying to run a pannel regression using the xtreg command with fixed effects (after performing a hausman test which proved to be significant). My command looks as follows:

    Code:
     xtreg erm_index_w L1.c.ICSR##L1.i.cso_presence c.emp_log c.leverage_w i.bigfour i.is_financial i.crisis_dummy, fe vce(robust)
    Note that xtset for each pannel id and year has been set.

    When performing this, i.is_financial gets ommited. The value does not change over firm or year, as it indicates whether the firm is part of a financial industry or not. How can I solve this issue? Are there any workarounds? Is another regression model more suitable, although the hausman test showed this one to be the best?

    Thank you!

  • #2
    Look into running a correlated random effects model, see Wooldridge (2010).

    Comment


    • #3
      Marie:
      welcome to this forum.
      As an aside to Maxence's helpful guidance, see The Stata Blog ยป Fixed effects or random effects: The Mundlak approach
      Kind regards,
      Carlo
      (Stata 19.0)

      Comment

      Working...
      X