Announcement

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

  • Consultation on regression model with dichotomous variable taken in a longitudinal period ( 2014-2018)


    Good evening I am new with STATA and I have a query, If I have a dichotomous outcome, but my study has a report for several years (annual period between 2014-2018). To look for associated factors, the entire population could use a logistic model adjusting according to the years to treat the variance that the years of study could possibly generate.

    I would appreciate your response.

  • #2
    You don't really pose a question, but maybe you're looking for something like xtgee with a suitable distribution family and link function?

    Comment


    • #3
      Thanks for your answer.

      What I'm looking for is to know if I have a period (2014-2018) with Unique IDs, that is, there is no tracking. I can take it as if it were a cross-section and use Logistic Model and without having to adjust that model according to the years. For example: I have a 4 year report of students who were violent

      Comment


      • #4
        Once again, you don't pose a question. Try having a statement that needs a question mark at the end of it.

        If you have panel data, then for the most part, you should use panel-based techniques like xtreg. Often, including year dummies (i.year) is important to pick up any general changes that might influence the dv so they're not mis-attributed to the included variables.

        Comment

        Working...
        X