<>
what about:
***
#delimit;
_pctile pincp [aweight=pwgtp] if agerange==1 & puma=="02200" & dropout==1,
nq(10);
sum pincp [aweight=pwgtp] if agerange==1 & puma=="02200" & dropout==1 &
pincp >= r(r1) & pincp < r(r2), de;
di "`r(p50)'";
di "`r(sum_w)'";
***
greets, rich
[email protected] schrieb am 24.09.2009 15:58:04:
> Yes - that works. Thanks. However, output generated after the sum
> command shows N to be my sample, and not the estimated population.
Thoughts?
>
> -----Original Message-----
> From: Nick Cox [mailto:[email protected]]
> Sent: Thursday, September 24, 2009 5:25 AM
> To: [email protected]
> Subject: st: RE: using xtile and qreg to find the median income for
> individuals between certain income percentiles
>
> Why not just use -summarize, detail- with -aweights-?
>
> Nick
> [email protected]
>
> Holly Kosiewicz, M.A.
>
> I have a database that contains information on a sample of individuals.
> The database includes a final weight that is used to produce estimates
> for the population.
>
> I have to used the xtile command to generate 10 percentiles for a
> variable that measures income conditioned on a group of individual
> characteristics: working age, dropout, and lives in a particular area;
> the exact command I use is:
> xtile ten=pincp [aweight=pwgtp] if agerange==1 & puma=="02200" &
> dropout==1,nq(10)
>
> I would like to find the weighted median income for individuals that
> fall between the 1-10th percentile, 11th-20th percentile, 21-30th
> percentile of the income variable.
>
> Martin, Nick, and Austin have shown me that the qreg command can be used
> to obtain the median income level for the entire population. I have
> tried to use this command to find the median income for individuals that
> fall between certain income percentiles, but Stata gives me an error
> stating that my income variable is dropped due to collinearity
> (something I should have expected from the get-go). This is the command
> I use: qreg pincp [aweight=pwgtp] if agerange==1 & puma=="02200" &
> dropout==1 & ten==1
>
> I have looked at different posts about topics dealing with percentiles
> and weights but none of them directly tackles this issue.
>
>
> *
> * 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/
>
>
>
> *
> * 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/
*
* 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/
