Re: st: Standard error of median using pweights

[Stas Kolenikov <skolenik@gmail.com>]

I've updated the package on my webpage, which should take care of the
first error. What's happening with the other one, I don't know. I
would suspect that you don't have enough observations in a cell, or
something of that kind.

On Tue, May 3, 2011 at 10:47 AM, Wassim Tarraf <tarrafwassim@gmail.com> wrote:
> Stas- I tried using your -epctile- package to calculate the median (please
> see below), but I keep getting error messages that I am unable to explain.
> Any guidance on this would be appreciated.
>
> svyset psu [pw=saqpw], strata(strat)
>
> *Median over x1 categories
>
> epctile y, svy percentiles(50) subpop(if x2==1 & x3==1) over(x1)
> unrecognized command:  FormXLevels
> r(199);
>
> *Median subpoping on x1 categories
>
> epctile y, svy percentiles(50) subpop(if x2==1 & x3==1 & x1==1)
>
> estimates post: matrix has missing values
> r(504);
>
> Thanks,
> Wassim
>
> On 5/2/2011 4:11 PM, Stas Kolenikov wrote:
>>
>> On Mon, May 2, 2011 at 10:29 AM, David Radwin<dradwin@mprinc.com>  wrote:
>>>
>>> Dear Stata-listers,
>>>
>>> Is it possible to calculate the standard error of a median using
>>> pweights,
>>> and if so, how? The dataset is a survey with a BRR design.
>>>
>>> This 2007 posting suggests it is possible but does not get into
>>> specifics:
>>>
>>> http://www.stata.com/statalist/archive/2007-03/msg00296.html
>>
>> I have a package that does this.
>>
>> net from http://web.missouri.edu/~kolenikovs/stata
>> net d epctile
>>
>>
>
> *
> *   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/
>



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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/