Re: st: RE: Idenity of entitiy at pctile value

[Rod Hunter <info@hunarch.com.au>]

Bingo! Thanks Matt.

Could you tell me what command I should use to identity the observation 
that is closest to a percentile value for a variable, and the 
corresponding values of other variables for that observation? I can get 
the answer, but only in a cumbersome way.


Rod Hunter

Schonlau, Matthias wrote:

> May be missing values are treated differently ? 
> Matt  
> -----Original Message-----
> From: Rod Hunter [mailto:info@hunarch.com.au]
> Sent: Thursday, January 30, 2003 4:18 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Idenity of entitiy at pctile value 
>
> A very rudimentary question:
>
> I want to identify the entity that corresponds with the 90th percentile 
> for area.
>
> I have performed _pctile, and generated a new variable based on the 
> result of _pctile. I have then used tabstat to identify the entity 
> corresponding with the p90 for area. That's fine. I get an answer, 
> including the value of p90 for area (902720)
>
> I also tried rtab, with the following results:
>
> ...
> 49.	909860	1	0.20	90.34
> 50.	908485	1	0.20	90.14
> 51.	906000	1	0.20	89.94
> 52.	902720	1	0.20	89.74
> 53.	901632	1	0.20	89.54
> ...
>
> The tabstat result of 902720 appears to be discrepant with the result of 
> rtab.
>
> I would be grateful if someone could explain.
>
> Many thanks
>
> Rod Hunter
> Hunarch Consulting
>
>
>
> *
> *   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/
> *
> *   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/

*
*   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/