The help for -makematrix- tells you why, although
you must read it very closely.
The outline syntax for -makematrix- is
makematrix [matrix_name] ,
from(results_list) [production_options] [list_options] :
["]command["] [varlist] ... [, options ]
In particular, following the colon : a command must be specified together
with whatever elements from
[varlist] [if exp] [in range] [weight] [, options ]
are appropriate.
Now that last syntax is fairly general, and covers lots of
commands. However, the -ttest- command you are using is not
among them, given the equals sign that is part of the syntax.
In short, -makematrix- is not general enough to allow what
you want to do. It falls over given the = sign.
As I have indicated before, -makematrix- was a project that
led to a program I thought worth publishing, but it's still
a dead end.
For what you want to do, a more basic approach is simpler
and more flexible:
ttest y = x
mat a = r(p) \ r(t)
mat rownames a = p t
Nick
[email protected]
Herve STOLOWY
> A few months ago, I asked a question concerning a way to get
> the t test results in a matrix. I received an answer based on
> -makematrix- (downloadable from SSC) and the following
> command works perfectly:
>
> makematrix c1, from(r(t) r(p)) label : ttest var1, by(groupvar)
>
> I am now running a different form of t test:
>
> makematrix c1, from(r(t) r(p)) label : ttest var1 = var2
>
> Unfortunately, I get an "invalid syntax" error message.
>
> Would you know why?
*
* 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/
