Re: st: Rolling regression without "clear"

["G. Anderson" <ga274@cam.ac.uk>]

Works well. Thank you!

On Jun 13 2013, Nick Cox wrote:

> There is no difference in principle. Even if you have a stored scalar, say
>
> scalar foo = 42
>
> you can supply its value like this
>
> ...  , window(`=foo')
>
> so that what happens is
>
> 1. Stata sees the expression to be evaluated `=foo'   -- goes then
> into a corner -- and comes back with the result of that evaluation,
> here 42
>
> 2. -rolling- itself sees -window(42)- and moves on.
>
> If your name could be confused with a variable name, then
>
> ..., window(`=scalar(foo)')
>
> disambiguates.  Documentation starts at -help macro-, but you have to
> dig a bit deep to see it from documentation.
>
> For a local macro,
>
> ..., window(`foo')
>
> would work, but for a scalar, a beast of a different kind, syntax is
> slightly different.
>
>
> Nick
> njcoxstata@gmail.com
>
>
> On 13 June 2013 11:33, G. Anderson <ga274@cam.ac.uk> wrote:
> > Great. Thanks Nick.
> >
> > Do you know whether it's possible to set the window size based on a 
> > stored scalar, rather than typing in the number?
> > Thanks,
> >
> > Gareth
> >
> >
> >
> > On Jun 13 2013, G. Anderson wrote:
> >
> > > Hi,
> > >
> > > I am using the "rolling" command in Stata 12 to do rolling
> > > regressions.However when I use
> > >
> > > rolling , window(150):regress depvar indvar
> > >
> > > I get the error message "no; data in memory would be lost"
> > >
> > > The only way I can get around this is by adding in the "clear" command,
> > > but this deletes all of my dataset, which is not ideal. Is there a way
> > > around this?
> > >
> > >
> > > Thanks,
> > >
> > > Gareth
> > > *
> > > *   For searches and help try:
> > > *   http://www.stata.com/help.cgi?search
> > > *   http://www.stata.com/support/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/