Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

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


From   "Gauri Khanna" <gwkhanna@hotmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   RE: st: RE: Frontier,stata 8 & stata 9
Date   Mon, 12 Jun 2006 08:26:36 +0000

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/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index