Hi,
I am working with scatterplots displaying the mean values of a
continuous dependent (y; income, for instance) for the different
levels of a discrete independent (x; years of education). The sizes
of the symbols in the plot are proportional to the group sizes, i.e.
to the numbers of observations, which have been used to calculate the
means of y.
I now would like to add a Lowess curve (or something alike) to the
plot to illustrate the functional form of the relationship between
the two variables.
I think that the the data points should be weighted according to
their group size. However, the -lowess- command does not allow
weights. Does anyone know a workaround (without having to use the
the individual level data)?
Thanks a lot for any help,
ben
*
* 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/
