Re: st: AW: can I get standardized coefficients by standardize variables first?

["Martin Weiss" <martin.weiss1@gmx.de>]

<>

BTW Mandy, as you say you want to continue with the "method for random 
insertion of missings" that I presented today, note that -runiform()- throws 
up a new vector of values every time the -foreach- loop is run, i.e. in this 
particular example five times. That means that a different 1% of 
rows -observations- is chosen to get a missing value. I am too tired now to 
think about the implications for your problem. But in the future you may 
want to construct the random choice of rows via

***********
tempvar unif
g `unif'=runiform()<0.01
foreach var of varlist fatal beertax spircons unrate perincK {
replace `var'=. if `unif'
}
***********

so that the same rows get hit...

HTH
Martin
_______________________
----- Original Message ----- 
From: "Martin Weiss" <martin.weiss1@gmx.de>
To: <statalist@hsphsun2.harvard.edu>
Sent: Tuesday, March 03, 2009 10:11 PM
Subject: Re: st: AW: can I get standardized coefficients by standardize 
variables first?


> <>
>
> So here is an implementation of Eva`s excellent advice as to the causes of 
> Mandy`s problem:
>
> *************
> clear*
> set mem 400m
> * enter the percentage of missings here
> loc mis 0.01
> u http://www.stata-press.com/data/imeus/traffic, clear
> foreach var of varlist fatal beertax spircons unrate perincK {
> replace `var'=. if runiform()<`mis'
> }
>
> qui xtreg fatal beertax spircons unrate perincK, fe
> tempvar dummy
> g `dummy'=e(sample)
>
>
> qui ds, not(type string)
> foreach var of varlist `r(varlist)'{
> egen std`var'=std(`var') if `dummy'
> }
>
> xi: reg fatal beertax spircons unrate perincK i.state if `dummy', beta
> xtreg stdfatal stdbeertax stdspircons stdunrate stdperincK if `dummy', fe
> *************
>
>
> HTH
> Martin
> _______________________
> ----- Original Message ----- 
> From: "Mandy fu" <mandy.fu1@gmail.com>
> To: <statalist@hsphsun2.harvard.edu>
> Sent: Tuesday, March 03, 2009 9:40 PM
> Subject: Re: st: AW: can I get standardized coefficients by standardize 
> variables first?
>
>
> > Dear Mr. Weiss,
> >
> > Thanks for providing the example data. I really  like the idea of
> > checking different percentage of missing to compare the results from
> > the two methods.
> >
> > This is what's shown when changing the percentage of missing:
> > -------------------------------------------------------------------
> > when missing percentage is zero:  coincide
> >
> > when missing percentage is 0.01: differences are very small
> >
> > when missing percentage is 0.01: differences are still small
> >
> > when missing percentage is 0.3: differences are obviously larger
> >
> > ----------------------------------------------------------------
> >
> > So, I'm thinking this shows well that  the missing values explains the
> > differences of the two methods. They give the same results when
> > adjusting the missing values.
> >
> > Thanks !
> >
> > Mandy
> > *
> > *   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/