Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: aweights for overall data when analyzing subsets of data
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: aweights for overall data when analyzing subsets of data
Date
Sat, 26 Mar 2011 00:33:22 +0000
What makes you say that? I did this experiment:
sysuse auto, clear
separate mpg, by(foreign) veryshortlabel
scatter mpg weight [w=price] , ms(Oh) mcolor(red) || scatter mpg?
weight [w=price] , ms(Oh ..) mcolor(blue ..)
Blue occludes red with no overlap or underlap on my graph.
(Note that -scatter- does not allow -iweights-.)
Nick
On Fri, Mar 25, 2011 at 7:55 PM, Laura Grant <[email protected]> wrote:
> I am scatter-plotting data for two subsets where I also care about a
> third dimension, such a population. I want to represent population
> using relative sizing of the symbols, by -aweight-
>
> However, running
>
> twoway (scatter y x [w= population] if use<50, msymbol(Oh)) (scatter
> y x [w= population] if use>=50, msymbol(Oh) )
>
> produces two subsets of the data in the same graph, but it splits up
> the weighting calculation so that the two sets' symbol sizes are NOT
> comparable. I would like to have the weights be over the entire data
> set. How? I think I should be able to -gen- the aweights as a variable
> then maybe tell Stata these are iweights?
>
*
* 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/