--- Fiona Wong wrote:
> When I use truncreg to estimate truncated regression, this error always
> appear:
> could not calculate numerical derivatives
> missing values encountered
> r(430);
>
> I tried to begin with very simple model: just use one independent
> variable, but this error still exists. My database has more than 20k
> data, this may cause problem to find initial value for the model. I
> also tried to take a sample from this database(1%), like 200, the
> model works. But I need to use all my 20k data, is there a better way
> to estimate truncated regression?
This could be due to outliers (which were by chance ommited in your
subsamples). This kind of technique can be very sensitive to outliers.
-- Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
