Re: st: using varlist wildcards in anova

[Phil Schumm <pschumm@uchicago.edu>]

On Apr 3, 2010, at 4:45 PM, Vineet Bhagwat wrote:
> So I guess I'm wondering why the command
>
> "anova y t* " won't work, it gives the error "t ambiguous  
> abbreviation"
>
> I created a smaller test dataset with only 3 variables, y, t1991,  
> and t1992 and get the same error, so I don't think it's stemming  
> from the dataset but how anova works.


Hmm, that's strange -- if you have only y, t1991 and t1992 in your  
dataset, the command

    anova y t*

should definitely work.  Of course, if you leave out the *, then  
you'll get what you described, i.e.,

    . anova y t
    t ambiguous abbreviation
    r(111);

You'll also get that if you have an errant space between the t and *:

    . anova y t *
    t ambiguous abbreviation
    r(111);

Are you typing the command interactively (i.e., in the command  
window), or are you executing it from a do-file?  Regardless, this  
isn't due to -anova-, but rather is a function of how Stata is parsing  
the command line.  For example, I presume that the command

    list y t*

will throw the same error for you?

If you want, take your dataset with 3 variables, type the commands

    keep in 1/3
    li
    anova y t*

which, according to what you've described should throw an error, and  
then copy the commands and output from the Results window and paste  
them directly in an email message.  One of us can then verify that we  
observe different behavior on our machine (i.e., that the command  
works); this will verify that you've got a problem with your Stata  
installation.


-- Phil

*
*   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/