Hi statalisters, am extracting the marginal effects from an ivprobit for
different values away from the mean in my main variable of interest. I obtain
unusually high effects (probabilities) at the bottom tail of the distribution
with probabilities beyond 2 at the very end and very high standard errors. I
know this is more an econometric question, but I wonder if am using the right
command and if so whether this sort of outcomes are possible and more
importantly, not signaling that some adjustments is necessary. Am using the
following command:
mfx compute, predict(p) at(mean X=...)
p% X MFX Std. Error
1% 0.05922 -2.280694 0.25423
5% 0.12922 -2.242234 0.5164
10% 0.17042 -2.053447 0.56453
25% 0.24695 -1.498927 0.46305
50% 0.34258 -0.7641651 0.2008
Mean 0.35219 -0.7024904 0.1774
75% 0.44465 -0.2669367 0.02902
90% 0.54960 -0.0629408 0.01264
95% 0.61274 -0.0220944 0.0094
99% 0.73249 -0.002104 0.00204
In a couple of instances (at low values of X) Stata also displays the following
warning: derivative missing; try rescaling variable X
Any guidance would be very mcuh appreciated.
Alejandro
--
Alejandro de la Fuente
Department of International Development/QEH
University of Oxford, Mansfield Road, Oxford OX1 3TB
Tel: 01865 281836
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
