Announcement

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

  • Different results: areg, cluster(id) absorb (id) and xtreg, fe vce (cluster id)

    Dear all,

    I've got another question concerning my panel regression (fixed effects model, xtreg,fe vce (cluster id)). Several sources state that it is more appropriate to report the R square of the "areg, cluster (id) absorb (id)" command instead of "xtreg,fe vce (cluster id)". However, both commands report different Standard Errors and different p values. Do you know why there's a difference? And is it really more appropriate to report the R squared provided by areg (besides the fact that it is higher and therefore more "favorable")?


    Thanks in advance!

    Ramona

  • #2
    Ramona:
    you can find the guidance you're looking for at http://www.stata.com/manuals13/rareg.pdf. (especially the Tecnical note at page 77).
    Kind regards,
    Carlo
    (Stata 18.0 SE)

    Comment

    Working...
    X