Re: st: RE: Syntax problems with maximum likelihood (method lf)

 From "Stas Kolenikov" To statalist@hsphsun2.harvard.edu Subject Re: st: RE: Syntax problems with maximum likelihood (method lf) Date Thu, 30 Oct 2008 08:47:27 -0500

```Adding to Glenn's comments about the problems with the code, it
appears that you need n to be defined externally -- as a variable or
as a scalar. Somewhere you have b without the localization embedders
(b instead of `b').

You might want to check out -mata optimize- instead of -ml-. The
former is a pretty universal optimizer, while the latter is really
intended to solve statistical problems with the data. If you want to
find the maximum/minimum of a function, it might be working out better
and result in a cleaner code with -mata optimize-.

On 10/29/08, Katia Bobulova <katia.bobulova@googlemail.com> wrote:
> Dear Nick,
>
>  I'll try to be more clear.
>
>  following my program:
>
>  program define myprogram
>  version 10.0
>  args lnf a b c e
>
>  tempvar sigma mc ma msigma mb
>
>  quietly gen double `sigma'=0.5*`e'
>  quietly gen double `mc'=1-`c'
>  quietly gen double `ma'=1-`a'
>  quietly gen double `msigma'=1-`e'
>  quietly gen double `mb'=1-`b'
>
>  quietly replace
>  `lnf'=ln((`a')*(`b')*((((`mc')*(`sigma')+(`m'))^alfa)*(((`mc')*(`sigma'))^s)*(((`mc)*(`msigma'))^n))+
>  (`a')*(`mb)*((((`mc') *(`sigma'))^b)*(((`mc') *(`sigma')
>
>  end
>
>  Everything is fine, but when I type:
>
>  ml model lf myprogram (a:) (b:) (c:) (e:)
>
>  Stata tells me that is invalid syntax.
>
>  What's wrong?
>
>  Thanks
>
>  Katia
>
>  2008/10/29 Nick Cox <n.j.cox@durham.ac.uk>:
>
> > It looks as if there is a bug within the part of the program you have not shown us.
>  >
>  > I suspect that you did not type
>  >
>  > version10.1
>  >
>  > as that would have generated a different error message.
>  >
>  > version 10.1
>  >
>  > is the correct syntax.
>  >
>  > In other words, you need to show more of your program to get better advice.
>  >
>  > Nick
>  > n.j.cox@durham.ac.uk
>  >
>  > Katia Bobulova
>  >
>  > I am using the likelihood function lf. I have a problem in the end. I
>  > have written the program and I don't have any problem.
>  >
>  > I typed:
>  > program define myprogram
>  > version10.1
>  > args lnf a b c d
>  > .
>  > .
>  > .
>  > end
>  >
>  > After this I tiped:
>  >
>  > ml model lf myprogram (a:) (b:) (c:) (d:)
>  >
>  > but Stata tells me that is an invalid syntax. Any suggestion?
>  >
>  >
>  > *
>  > *   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/
>

--
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.
*
*   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/
```