[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: marginal effects
I am trying to estimate some marginal effects after having estimated a model using the reoprob command (random effects ordered probit).However, as you may know in an ordered probit model you have more that one threshold to estimate.
When I use the following command: mfx compute, predict(outcome(eq1)) nose, Stata gives the change of the covariates from 0 to 1. In my model though I have 4 different thresholds. My question is how can I estimate the marginal effects of the covariates for the other thresholds? Do I have to specify the thresholds as different equations?
The above command: mfx compute, predict(outcome(eq1)) nose, works perfectly well after estimating a single ordered probit.
Thanks in advance for any suggestions,
* For searches and help try: