Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Re: optimization problem in Stata?


From   Oleksandr Shepotylo <shepotil@econ.umd.edu>
To   is300gp <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Re: optimization problem in Stata?
Date   Thu, 13 May 2004 00:56:06 -0400

Yes, matlab is very convinient for such kind of problems.

But I have 3-step-procedure to estimate coefficients. First and third  of
which are standard regressions. But second stage is some kind of
non-linear least squares and requires minimazing quadratic form.

Wednesday, May 12, 2004, 7:33:18 PM, you wrote:

i> Maybe you should try matlab or gauss.


i> ----- Original Message ----- 
i> From: "Oleksandr Shepotylo" <shepotil@econ.umd.edu>
i> To: <statalist@hsphsun2.harvard.edu>
i> Cc: "Alexand Shepotilo" <shepotil@econ.bsos.umd.edu>
i> Sent: Wednesday, May 12, 2004 11:36 AM
i> Subject: st: optimization problem in Stata?


>> Hi statalist,
>>
>> I am a little bit confused by optimization capabilities of Stata.
>>
>> Suppose I want to estimate parameters of a vector g=[alpha, beta]
>> by minimizing the following expression:
>>
>> min [A*g+B]*[C*g+D],
>>
>> where A,B,C,D are matrixes. What command of Stata if any is capable of
i> doing
>> this?
>>
>>
>> Thanks, Oleksandr.
>>
>>
>> *
>> *   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/
>>

i> *
i> *   For searches and help try:
i> *   http://www.stata.com/support/faqs/res/findit.html
i> *   http://www.stata.com/support/statalist/faq
i> *   http://www.ats.ucla.edu/stat/stata/



-- 
Best regards,
 Oleksandr                            mailto:shepotil@econ.umd.edu


*
*   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