Prev: Return the name of the file currently open in the Enhanced
Next: Return the name of the file currently open in the Enhanced Editor
From: Dale McLerran on 5 Mar 2010 13:26
--- On Thu, 3/4/10, Pakcha Sharma <pakcha(a)GMAIL.COM> wrote:

> From: Pakcha Sharma <pakcha(a)GMAIL.COM>
> Subject: Quantile regression
> To: SAS-L(a)LISTSERV.UGA.EDU
> Date: Thursday, March 4, 2010, 5:43 PM
> Hi all,
>
> I am trying to do quantile regression with data from a survey. Each
> record has sampling weight to represent the whole population. I am
> just wondering if PROC QUANTREG can be used for the survey data. Does it
> compute the correct standard error for the estimates if I specify
> sampling weights through the WEIGHT statement ? Or is there a problem
> like in PROC LOGISTIC, which does not compute the proper variance when
> using with the survey data and specifying sampling weights through the
> WEIGHT statement, and we need to use PROC SURVEYLOGISTIC instead of PROC
> LOGISTIC.
>
> Is there any other PROC that can be used for quantile regression for
> survey data ? Any help would be much appreciated.
>
> Thanks.
>
> Packha.
>

Packha,

The weight statement of PROC QUANTREG does not specify sampling
weights. Rather, the weights specified in PROC QUANTREG specify
only a function of the variance of the response. I don't know
of any procedure in SAS which can be used to perform a quantile
regression with adjustment of standard errors to account for
sample design considerations.

Dale

---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra(a)NO_SPAMfhcrc.org
Ph: (206) 667-2926
Fax: (206) 667-5977
---------------------------------------
 | 
Pages: 1
Prev: Return the name of the file currently open in the Enhanced
Next: Return the name of the file currently open in the Enhanced Editor