---- Yoshiro Nagao wrote:
> I am thinking of using Lowess smoothing
> for such a dataset in which
> each data-point corresponds to one community.
> Since different communities differ in
> population size,
> they should be weighted, as I suppose,
> according to their size.
---- Austin Nichols responded:
> Depending on what you actually want to do, you
> may find it helpful to -expand- your data by
> population size (the equivalent of fweights) and
> then run -locpoly- (local polynomial regression)
> on the expanded data.
Yoshiro:
When I had to do something like that I was unable
to find how to do that in Stata and I eventually
did it in R using the locfit package. (My weights
were inverse variances so I would not know how to
expand that and use -locpoly- with that)
HTH,
Maarten
http://www.r-project.org
http://www.locfit.info/
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z214
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
