Glenn thank you very much. I will follow your suggestion and try that
out soon.
Jinhu
On Fri, 28 Aug 2009 11:43:35 +0100
Glenn Goldsmith <[email protected]> wrote:
> <>
>
> >From your first post, the equation name is a1, so yes, you should
> use that:
>
> constraint 1 [a1]_cons=1.45
>
> Note that:
>
> constraint define 1 [a1]_cons=1.45
>
> should also work. The problem with your earlier code was that you had
> "1 define", rather than "define 1". This would have defined your
> constraint 1 as "define [eq2]_cons=1.45", which doesn't make sense.
>
> See also: http://www.stata.com/support/faqs/stat/constraints.html
>
> HTH
>
> Glenn.
>
> "J. Li" <[email protected]> wrote:
>
> Thanks Michael. I will try that out. I don't know if that could
> change
> the result though. I am not sure about the [eq2] part neither. But
> given that in the ml program, I use mleval to obtain the intermediate
> variables from the parameters as following:
>
> mleval `mu' = `b', eq (1)
> mleval `a1' = `b', eq (2) scalar
> mleval `a2' = `b', eq (3) scalar
> mleval `a3' = `b', eq (4) scalar
> mleval `a4' = `b', eq (5) scalar
>
> So I suppose when I want to contrain the parameter of the first
> cut-off
> value as 1.45, I should specify the equation name as eq2 so it is
> consistent with the name used in "mleval". Is this right? Or should I
> use
>
> constraint 1 [a1]_cons=1.45
>
> instead? Thanks!!
>
>
> Jinhu
> *
> * 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/
*
* 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/