Announcement

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

  • xttobit and margins, predict(e(0,.))

    Hello

    I have a panel data, and I have recently run a tobit model with random effects with the command xttobit. I get that almost all the estimators are significant, but when I want to calculate the marginal effects with "margins, dydx(*) atmeans predict(e(0,.))", the covariates are no longer significant and all p-values are "0.957". So, this fact (all p-values are 0.957) is very strange for me. So, I don't know what is happening.

    PS: I understand that commands written above calculate the marginal effects when dependent variable is left-truncated and I want to know E(y|y>0).
Working...
X