Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Saving svy mean estimates?


From   Matt Spittal <Matt.Spittal@cancervic.org.au>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Saving svy mean estimates?
Date   Wed, 08 Apr 2009 11:24:01 +1000

Rebecca,

Others are more proficient than me at using the matrix functions.  But the
principle to extract standard errors is the same as the means, but requires
a little extra work.  The variance-covariance matrix is stored in -e(V)-,
and we need the square root the diagonal elements to get the SEs.

    webuse highschool, clear
    svyset county [pweight=sampwgt], strata(state) vce(linearized)
    svy: mean weight, over(sex)

    matrix V = e(V)             // <- gets the variance-covariance matrix
    matrix Var = vecdiag(V)     // <- gets the diagonal elements

    svmat Var      
    replace Var1 = sqrt(Var1)   // <- take the square root to get SEs
    replace Var2 = sqrt(Var2)

No doubt there is a more efficient method to get the square root of the
diagonal elements. Others may have some thoughts on this.

-- Matt
matt.spittal@cancervic.org.au




On 8/4/09 10:52 AM, "Kanter, Rebecca" <rkanter@jhsph.edu> wrote:

> Thanks again everyone...
> 
> Now this may be asking too much, but following the below example of saving the
> matrix of mean values to variables...does anyone know how to save their
> corresponding standard errors (SEs) as a matrix and/or variable(s)? Thanks.
> [prior example:]
>     webuse highschool, clear
>     svyset county [pweight=sampwgt], strata(state) vce(linearized)
>     svy: mean weight, over(sex)
>     matrix means =  e(b)
>     matrix list means
>     svmat means     // <-- This line creates two new variables
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu
> [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Matt Spittal
> [Matt.Spittal@cancervic.org.au]
> Sent: Tuesday, April 07, 2009 7:46 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: Saving svy mean estimates?
> 
> Rebecca,
> 
> You can use the -svmat- command to create a variable from a matrix.  From
> the examples given previously
> 
>     webuse highschool, clear
>     svyset county [pweight=sampwgt], strata(state) vce(linearized)
>     svy: mean weight, over(sex)
>     matrix means =  e(b)
>     matrix list means
> 
>     svmat means     // <-- This line creates two new variables
> 
> I hope this is helpful.
> 
> -- Matt
> matt.spittal@cancervic.org.au
> 
> 
> 
> 
> 
> 
> On 8/4/09 10:30 AM, "Kanter, Rebecca" <rkanter@jhsph.edu> wrote:
> 
>> Joao,
>> Thanks for your response; that helps a bit. But what I'm trying to get at is
>> how to store the e(b) (i.e. mean values) into a variable [the same way as if
>> I
>> did egen newvar=mean(var), without acct for svy design]. Do you or someone
>> else know how to do this?
>> Many thanks,
>> Rebecca
>> ___________________________________________
>> Rebecca M. Kanter
>> PhD Candidate
>> Johns Hopkins Bloomberg School of Public Health
>> Department of International Health
>> Center for Human Nutrition
>> ________________________________________
>> From: owner-statalist@hsphsun2.harvard.edu
>> [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Joao Ricardo F. Lima
>> [jricardofl@gmail.com]
>> Sent: Tuesday, April 07, 2009 7:13 PM
>> To: statalist@hsphsun2.harvard.edu
>> Subject: Re: st: Saving svy mean estimates?
>> 
>> Rebecca,
>> 
>> the svy mean estimates are saved in e(b)
>> 
>> *********begin example*************
>> webuse highschool, clear
>> svyset county [pweight=sampwgt], strata(state) vce(linearized)
>> svy: mean weight, over(sex)
>> eret li
>> mat li e(b)
>> *********end example*************
>> 
>> Sorry but I didn´t understand the second part of your question.
>> 
>> HTH,
>> 
>> Joao Lima
>> 
>> 2009/4/7 Kanter, Rebecca <rkanter@jhsph.edu>:
>>> Hi,
>>> 
>>> I am using the svy mean command to calculate survey weighted state-specific
>>> mean estimates of different Mexican state characteristics, for example the
>>> percent of the population in a specific Mexican state that is indigenous
>>> 
>>> Thus, I use the following command line: svy: mean indigena, over(state)
>>> 
>>> And want to save these mean estimates (by state) as I would if I did an by
>>> state, sort: egen mean(indigena)--that was not accounting for the survey
>>> design; so my question is, how do I save the mean estimates by state from
>>> the
>>> above svy command line: svy: mean indigena, over(ent)??  And/or how do i
>>> save
>>> mean estimates for this that does take the svy pweight into account in the
>>> proper manner (if I can't do it with svy: mean)? [I want to then drop people
>>> from my dataset but want to have all the state estimates of characteristics
>>> saved as variables before doing so].
>>> 
>>> Thanks,
>>> Rebecca
>>> 
>>> 
>>> ___________________________________________
>>> Rebecca M. Kanter
>>> PhD Candidate
>>> Johns Hopkins Bloomberg School of Public Health
>>> Department of International Health
>>> Center for Human Nutrition
>>> 
>>> *
>>> *   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/
>>> 
>> 
>> 
>> 
>> --
>> ----------------------------------------
>> Joao Ricardo Lima, D.Sc.
>> Professor
>> UFPB-CCA-DCFS
>> Fone: +5538387264913
>> Skype: joao_ricardo_lima
>> ----------------------------------------
>> 
>> *
>> *   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/
> 
> *
> *   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index