st: Re: r(estimate) in svymean.

["Michael Blasnik" <michael.blasnik@verizon.net>]

svymean is an eclass command and leaves it's results in e().  Type -ereturn 
list- instead.  The r-class results are probably left behind by a command 
executed within svymean.

Michael Blasnik
michael.blasnik@verizon.net


----- Original Message ----- 
From: <adiallo5@worldbank.org>
To: <statalist@hsphsun2.harvard.edu>
Sent: Monday, December 06, 2004 9:34 AM
Subject: st: r(estimate) in svymean.



> Hi,
> I want to save the results produced by svymean. But I get confused by the
> saved results.
>
> . svyset [pweight=we] ,  psu(a1)
> pweight is we
> psu is a1
>
> . svymean g1_1
>
> Survey mean estimation
>
> pweight:  we                                      Number of obs    = 
> 270805
> Strata:   <one>                                   Number of strata = 
> 1
> PSU:      a1                                      Number of PSUs   = 
> 3267
>                                                  Population size  = 
> 1.705e+08
>
> ------------------------------------------------------------------------------
>    Mean |   Estimate    Std. Err.   [95% Conf. Interval]        Deff
> ---------+--------------------------------------------------------------------
>    g1_1 |   .0289745    .0033472    .0224118    .0355373    107.8358
> ------------------------------------------------------------------------------
>
> . ret li
>
> scalars:
>                 r(N) =  270805
>          r(N_strata) =  1
>             r(N_psu) =  270805
>             r(N_pop) =  270805
>          r(estimate) =  .034090951053341
>               r(Var) =  1.21596276679e-07
>           r(Var_srs) =  1.21596276678e-07
>           r(errcode) =  0
>
> My question is why r(estimate) (.034090951053341) is so different of the
> estimate displayed in table (.0289745) ?
> I try this with different variables and datasets and get the same problem.
> Regards.
> Amadou.
*
*   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/