I want to estimate the following:

$$Y = b_0 + b_1D_1 + b_2X_1 + b_2X_1D_1 + b_3X_3....+error$$

where $Y$ = outcome, continuous
$D_1$ = endogenous, binary
$X_1$ = exogenous, binary
$X_1D_1$ = interaction of the two
$X_3, X_4$ = exogenous variables

I have a panel data. From what I read so far, though not finding something that answers directly my question, I seem to be needing to do something like:

  • run an xtlogit $D$ $X_3$ $X_4$ etc
  • predict $\hat{d}$, pr
  • use as instruments for the interaction $\hat{d}$ and $Z\times \hat{d}$ , where $Z$=instrument, continuous.

Any comments or ideas about this?

  • $\begingroup$ Why do you need a panel ordered logit if the endogenous variable is is binary? $\endgroup$
    – dimitriy
    Jan 10, 2017 at 23:21
  • $\begingroup$ Sorry there was a typo. It was just a logit not an ordered one. $\endgroup$
    – Rebecca
    Jan 11, 2017 at 12:42


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.