Adapting the Many Tests of Inequality syntax for multiple zip codes

[Laflamme, David]

Hi Folks,

Has anyone used the inequality measures syntax at the link below to
calculate the Gini Index at the zip code level?  I would like to end up
with a single file showing each zip and the corresponding Gini
coefficient.  I don't care about the rest of the inequality measures or
producing the Lorenz curve.

Because it is a fairly long program, I thought it better to provide the
link.  Happy to post it if preferred.  Credit for the program goes to
Raynald Levesque.
http://www.spsstools.net/Syntax/Inequality/ManyTestsOfInequalityV5.txt

I am also unsure of what the best way is to structure the data for the
calculations.  At present, I have it in this format (because it works
when there is only one zip in the data file:

Group     a          e
zip1     89        5000
zip1     109       12500
zip1     60        17500
zip1     56        22500
zip1     87        27500
zip1     49        32500
zip1     90        37500
zip1     99        42500
zip1     110       47500
zip1     101       55000
zip1     70        67500
zip1     69        87500
zip1     65        112500
zip1     50        137500
zip1     45        175000
zip1     21        389764
zip2     50        5000
zip2     66        12500
zip2     67        17500
zip2     70        22500
zip2     88        27500
zip2     90        32500
zip2     98        37500
zip2     101       42500
zip2     109       47500
zip2     111       55000
zip2     94        67500
zip2     88        87500
zip2     70        112500
zip2     51        137500
zip2     43        175000
zip2     19        465600

Where "Group" is the zip code, and "a" is the number of people in the
corresponding zip code making "e" dollars.  Note that the values of "e"
are the same for each zip code with the exception of the last value for
each zip.  The matching values in "e" represent the mid-points of the
income categories used by the census.  However, the last value is the
mean income for people making over $200,000 -- which varies by zip code.

I can successfully run the program for a single zip code, but don't yet
have the programming skills to adapt it to run on all of them at once
and produce a single file showing the Gini for each zip.  I am hoping
that someone might have done this before and can share the adapted
syntax, or at least give me an idea of what steps would be necessary to
adapt the program myself.  Thanks for whatever advice you can provide!

~David