Announcement

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

  • IPWRA with independent categorical variable # less than 10 obs


    Hi Statalisters,

    I am running `teffects ipwra` for causal inference, but the regression does not converge. I found that if I remove a categorical variable, it converges.

    Upon inspection, this categorical variable has 30 levels, many of which have fewer than 10 observations.

    I must not combine some of the levels of this categorical variable. Moreover, I can't exclude if from the regression.

    Can anyone suggest how to proceed? Propensity score matching is essential for my analysis.

    Thanks.



Working...
X