> 1. Is it possible to apply PSPLINE to panel data?
No. Currently not. But I guess it should be possible. I'll look into
that (but cannot make any promises). It would be a model with crossed
factors.
> 2. How can one obtain confidence intervals for the splines after PSPLINE?
This is currently not possible because -predict- cannot obtain the
standard errors for the random effects after -xtmixed-. It is planned
to add a -ci- option to -pspline- once standard errors for random
effects become available in Stata (but I don't know when this will
happen).
ben
On Fri, Jan 30, 2009 at 9:29 AM, Christoph Wunder
<[email protected]> wrote:
> Two short questions regarding PSPLINE:
>
> 1. Is it possible to apply PSPLINE to panel data? I suppose that XTMIXED has
> to be invoked with an individual specific random error term added.
>
> 2. How can one obtain confidence intervals for the splines after PSPLINE?
>
> Any suggestions welcome. Thank you!
> Christoph
>
> *
> * 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/
>
*
* 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/
