RE: st: RE: Frontier,stata 8 & stata 9

["Gauri Khanna" <gwkhanna@hotmail.com>]

Dear Scott,

Thank you for your advice. I will look into the maximise options and see if 
it is possible to use another.

On the noconstant option, I wanted the option of not including it as for 
certain models it was not significant. I don't think that the stata versios 
is updated which might be part of the problem as you suggest.

Regards,

Gauri



> From: "Scott Merryman" <smerryman@kc.rr.com>
> Reply-To: statalist@hsphsun2.harvard.edu
> To: <statalist@hsphsun2.harvard.edu>
> Subject: RE: st: RE: Frontier,stata 8 & stata 9
> Date: Sat, 10 Jun 2006 13:17:32 -0500
>
>
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> > statalist@hsphsun2.harvard.edu] On Behalf Of Gauri Khanna
> > Sent: Saturday, June 10, 2006 1:34 AM
> > To: statalist@hsphsun2.harvard.edu
> > Subject: RE: st: RE: Frontier,stata 8 & stata 9
> >
> > Dear Scott,
> >
> > Thank you for your response.
> >
> > I tried the difficult option before but it had not worked. I even looked
> > at
> > the gradient vector to see where it nears zero. It does near zero after
> > several iterations and i thought that since the gradient does not change
> > perhaps I can take that as a cue to stop the iterations at that point.
> > Alternatively, I could arbitrarily set the iterations to a maximum of 
> say
> > 500 and then take the values at that point.
> >
> > Regarding the maximsation problem, I am not aware of this.. where can i
> > get
> > more information on this?
>
> Take a look at -help maximize-.
>
> "technique(algorithm_spec) (ml-programmed estimators only) specifies how 
> the
> likelihood function is to be maximized.  The following algorithms are
> currently implemented in ml.  For details, see Maximum Likelihood 
> Estimation
> with Stata, 2nd edition (Gould, Pitblado, and Sribney 2003).
>
> technique(nr) specifies Stata's modified Newton-Raphson (NR) algorithm.
>
> technique(bhhh) specifies the Berndt-Hall-Hall-Hausman (BHHH) algorithm.
>
> technique(dfp) specifies Davidon-Fletcher-Powell (DFP) algorithm.
>
> technique(bfgs) specifies the Broyden-Fletcher-Goldfarb-Shanno (BFGS)
> algorithm.
>
> The default is technique(nr)...."
>
> In couple examples I have tried, it take much longer to converge when the
> constant is excluded from -uhet()-.  Are there strong reasons for excluding
> it?
>
>
> > I guess you do not know have an answer for the stata 8 question where i
> > get
> > a "invalid noconstant" response to my equation:
> >
> > frontier logoutput logcorarea  loglabour logmanure logfertiliserval
> > logtrac
> > logox logirrigation cropdum, uhet(experience age corarea, noconstant).
> >
>
> I no longer have Stata 8.2 installed and the noconstant option works fine 
> in
> version 9.2.  Is your Stata fully updated?
>
> Scott
>
>
>
> *
> *   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/
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

*
*   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/