I agree with Scott. -xtline- is precisely the
tool designed for the purpose. If you want something
slightly different, try also -linkplot-
from SSC.
If Allan is right in his guess that Timothy
is holding data in wide form, then the advice
to -reshape- is also good. That's also tacit
in the advice to use -xtline-, so there's no contradiction
here really.
That said, users re-creating graphs for such
data from first -graph- principles are likely to be bit
by a problem of spurious connections joining together
curves from two or more individuals whenever
the last data point for one person is no
later than the first data point for the next person.
This may be rare, but it does occur. The graphical problem
is soluble, but fiddly.
So in practice -- as the existence of FAQs and previous
programs tackling this problem attest -- customised programs
get you there less painfully.
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of
> Allan Reese
> Sent: 19 January 2004 11:02
> To: Stata distribution list
> Subject: Re: st: Re: Plotting Growth Curves
>
>
> > From: "Timothy W. Victor" <[email protected]> asked
> > Sent: Saturday, January 17, 2004 11:55 PM
> > >
> > > I have repeated measures on 604 subjects. I'd like to
> get a visual feel
> > > for the functional form of the growth curves for this
> sample. Is there
> > > an easy way to plot all curves for all subjects (or
> most) in one graph?
>
> and On Sun, 18 Jan 2004, Scott Merryman wrote:
> > Try -xtline-
>
> That would work, but smacks somewhat of the "separate tool
> for every task"
> approach. Hence I think it worth mentioning another method
> which is to
> reshape the data from wide (the repeated measures form) to
> long (factorial
> observation form). The time point then becomes a new
> variable and can be
> used as X in graphs. Without further reading and tests, I
> do not know the
> advantages and drawbacks of either method. I doubt that
> any graph with
> 604 lines on it would be readily interpreted, but Stata has
> the power to
> sample or select cases.
>
*
* 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/
