Announcement

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

  • Help! Xtoprobit - panel standard deviation

    Good afternoon to everybody. I was estimating a regression through xtoprobit command and observed that the estimation also gave me the parameter "sigma_u" which I found reading the Stata help was the "estimated panel-level variance component"

    My question was about the interpretation of it. How may I interpret the results of this estimator? Whats would be the interpretation if it is significant at all levels and equal to zero?

    Thanks in advance!

    Angelo

  • #2
    Angelo:
    what does the LR test at the foot of the results table tell you?
    As an aside (echoing the FAQ) your chances of getting helpful replies is conditional on posting exactly what you typed and what Stata gave you back. Thanks.
    Kind regards,
    Carlo
    (Stata 18.0 SE)

    Comment


    • #3
      Thanks lor your answer Carlo, and Im sorry for not posting the commands and the output. Here the command for the regression where
      Code:
      $Lxhog
      is the global which defines the vector of independent variables.

      Code:
      xtoprobit transicion $Lxhog i.estrato, vce(cluster conglome) difficult  nolog /*gradient iter(100)*/
      estimates store vart
      When I run this regression i get this results (sorry, uploading a pic was the easiest way to show I found)


      Click image for larger version

Name:	Untitled.png
Views:	2
Size:	164.1 KB
ID:	1308316






      Comment


      • #4
        Angelo:
        the cause of your problem here seems that you have too few observation per cluster.
        Kind regards,
        Carlo
        (Stata 18.0 SE)

        Comment


        • #5
          Thanks Carlo! Do you think this result can be interpreted in any way? I am u sing this cluster corrected errors because I have a complex survey with this kind of design.

          Comment


          • #6
            Angelo:
            first off, I would skim the existing literature in your research filed and see whether others did a regression with less predictors in dealing with the same research topic.
            With such a limited average number of predictors per group, you run the risk of overfitting your model.
            Kind regards,
            Carlo
            (Stata 18.0 SE)

            Comment

            Working...
            X