Re: st: Using scalars when defining matrices

[Rachel <academicgirl@gmail.com>]

Thanks for all your replies.  I have tried several solutions,
including using local variables, but nothing has worked.  The relevant
lines of my ado file are as follows:

1:arg {other vars} abc
2: tempname beta BigMat
3 scalar `beta'=exp(`abc')
4 matrix input `BigMat'=(1,`scalar(beta)'\`scalar(beta)',`scalar(beta)')

When I trace the program, stata is reading line 4 as

"= matrix __00001A=(1,,,\ ,1,0,0"  and gives me the error message "1, not found"

The problem, I suppose, is that I'm not clear on how to use scalars in
conjunction with tempnames.  I have read the [P] manual on scalars,
but they are not specific on this point.

(I have tried various other ways to write line 4 (i.e. scalar(`beta')
instead of `scalar(beta)').  I"ve also tried to define `beta' as a
local

scalar `beta'=exp(`abc')

but this is obviously problematic because it evaluates `beta' as the
string "exp(`abc').)

Thanks again for all your help.




On 12/7/06, Michael S. Hanson <mshanson@mac.com> wrote:
> On Dec 7, 2006, at 7:48 PM, Rachel wrote:
>
> > I'm having trouble with the syntax of matrix define (or input) when
> > using scalars.  When I try the following:
> >
> > scalar b3=2
> > matrix input Matrixnew=(1,`b3',`b3',`b3'\ `b3',1,0,0\ `b3',0,1,0\
> > `b3',0,0,1)
>
>         [snip]
>
> > What am I doing wrong?
>
>         You are confusing scalars with local macros.  Consider the following:
>
> scalar b3 = 2
> local b3 = 4
> matrix Ms = (1,b3,b3,b3 \ b3,1,0,0 \ b3,0,1,0 \ b3,0,0,1)
> matrix Ml = (1,`b3',`b3',`b3' \ `b3',1,0,0 \ `b3',0,1,0 \ `b3',0,0,1)
> mat list Ms
> mat list Ml
>
>         Notice that b3 -- the scalar -- and b3 -- the local macro -- are two
> *different* objects with two *different* values.  That may be confusing
> to you, but Stata understands the difference.  And it's important it
> does, otherwise programs (do and ado files) could accidentally clobber
> a variable, scalar, matrix, etc. that might exist in the user's active
> memory prior to a program being executed if it just so happened that
> they had the same name.
>
>         As long as you are consistent -- that is, use a scalar and do not
> quote it, or use a local macro and quote it appropriately -- you likely
> will not get confused either.  I find that using scalars is more
> convenient when working interactively in Stata, but for any kind of
> programming (i.e. do or ado files), local macros are well worth
> learning and using.  The manuals are the best place to learn this topic
> -- far more complete than the on-line help.
>
>                                          -- Mike
>
> *
> *   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/
*
*   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/