Announcement

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

  • About insuffucient observeation error in doing with panel data

    Dear all
    I am trying to run Green's 2005 true random effect model. In doing this I declared my panel data using the following command and run the subsequent initial regression.
    xtset uid year
    quietly xtreg lnTR_ETB $inputs
    The uid is unique for each observation (5707). But the resultant outcome is an error message (insufficient observation). I have checked for missing variable and there is nothing.
    Anyone who can advise me to sort out the problem ?
    Thank you

  • #2
    You can't use -xtreg, re- in data where the panel variable uniquely identifies observations. Here's why:

    -xtreg, re- first tries to estimate -xtreg, fe- and -xtreg, be- and then calculate a weighted average of the results of those.
    If the panel identifier uniquely identifies observations, then there is only 1 observation per panel. But -xtreg, fe- is a within-panel estimator, so if there is only one observation per panel, within-panel estimation is impossible. It is this -xtreg, fe- that is giving the "insufficient observations" error message. And once that happens, Stata stops.

    Actually, apart from Stata's particular way of doing -xtreg, fe-, the whole idea of doing a random effects regression with only one observation per panel makes no sense. If the model is y = b0 + b1x1 + b2x2 + ... + bnxn + u + e, with u being the panel random effect and e being the observation level error, it is clearly impossible to separately identify u and e if there is only one observation per panel.

    To do this, you need data with more than one observation per panel.

    Comment


    • #3
      Thank you very much Clyde!!! I have sorted out that and I prefer to use the following command, which accounts heteroscedasticity. sfpanel lnTR_ETB $inputs, model(tre) distribution (hnormal)usigma () vsigma () nsim(50)simtype(genhalton) base (7) rescale But I am not clear that which variables are to be included in vsigma () and usigma () and how to select those variables. Thank you!!!

      Comment


      • #4
        I am not familiar with the -sfpanel- command. Perhaps somebody who is can answer your question.

        Comment

        Working...
        X