SR Millis <[email protected]> asks how to get -scatter- in Stata 8 to do
what pre-stata-8 graphs did with iweights,
> In Stata 7, after fitting a logistic regression model, I could
> produce a plot of H-L dx2 vs. the estimated probability with the
> size of the plotting symbol proportional to Pregibon's dbeta using
> iweights:
>
> . graph dx2 p [iweight=db]
>
> Stata 8's scatter won't accept iweights. Is there another way to
> produce this plot?
Scott Merryman <[email protected]> suggested using aweights, and I agree.
Or, use pweights or fweights; all three will produce the same graph.
-iweights- are no longer allowed, and never should have been, because iweights
can be negative and -scatter- does not know what to do with negative weights.
-- Vince
[email protected]
*
* 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/
