Re: st: sample size estimates for multilevel models

[Richard Goldstein <richgold@ix.netcom.com>]

Thanks but ...

Although I bought the biggest Mac available, and although I have 
attached external hard drives, and will need to attach some more, I 
still don't really want to give up any room to any version of Windows -- 
esp. since my dislike of Windows and its instability were what moved me 
to the Mac

If I have to do that I will -- but I definitely don't want to do that

Rich

David Airey wrote:
> .
>
> I run software on Windows on my Mac as needed, like SAS. Works fine with 
> Parallels 4 and XP or Vista. If you still have your valid licenses, why 
> not use them?
>
> -Dave
>
> On May 13, 2009, at 12:02 PM, Richard Goldstein wrote:
>
> > Having recently switched from Windows to a Mac, I find myself missing 
> > one thing from the Windows world -- software for sample size 
> > estimation for multilevel models (models that one would use, e.g., 
> > -xtmixed- or -gllamm- for).
> >
> > I find that, within Stata, I can get what appears to be a reasonable 
> > estimate by -sampsi- for repeated measures and following this with 
> > Joanne Garrett's -sampclus- (from STB 60).  However, I would like to 
> > check the reasonableness of the answer if possible either within Stata 
> > or with Mac software (if any).
> >
> > Any suggestions are very welcome.
> >
> > Thanks,
> >
> > Rich
*
*   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/