Announcement

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

  • Problem with margins in multilevel random slope and stata

    I run a multilevel model
    xtmelogit vd vi if eurozone==1 || countryid:, ml var
    and the results come out with the independent variable as statistically significant.

    If I repeat the same multilevel model but adding different slopes for the same variable that was significant before
    xtmelogit vd vi if eurozone==1 || countryid: vi, ml var
    the v1 loses its significance because the standard deviation is very large.

    Does this make sense? Should I keep the first model or the second one?

    I would like to represent the margins of an interaction in the model
    xtmelogit vd vi##vi2 if eurozone==1 || countryid: vi, ml var
    How can i do it?

    Stata only allows me to represent the fixed part when I make margins.

    If not a get an error.
    margins, dydx(vi2) at(v1=(5(10)20)) predict (mu fixed)
    Thank you.

  • #2
    Your post is very confusing and omits crucial information needed to answer your questions..

    You keep referring to things happening to a variable v1, but your model does not include any such variable.

    For help understanding the differences in the outputs of two models, you need to show the outputs themselves. Please use code delimiters when posting model output so it aligns in a readable way. If you are not familiar with code delimiters, read Forum FAQ #12.

    If you want help troubleshooting an error message you are getting, you need to show the error message itself, as well as the command that generated it.

    Comment

    Working...
    X