Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

Re: st: question about Weighted Least Square(WLS)

 From Wu Zhang To "statalist@hsphsun2.harvard.edu" Subject Re: st: question about Weighted Least Square(WLS) Date Tue, 2 Apr 2013 17:28:05 -0700 (PDT)

```Hi Rich,

That really helps. Thank you very much!
By the way,  I could run the WLS as the GLS like this:
gen xstar=x/sqrt(exp(v))
gen ystar=y/sqrt(exp(v))
reg ystar xstar

So the parameter estimates should be the same as  method 1 and 2, right? But in my case, they are different.

P.S.:
> 1.  regress y x [aweight = 1/exp(v)]
>
> 2. vwls y x , sd(sqrt(exp(v)))

I am very confused .

Thanks again,

Wu

________________________________
From: Richard Goldstein <richgold@ix.netcom.com>
To: statalist@hsphsun2.harvard.edu
Sent: Tuesday, April 2, 2013 4:16 PM
Subject: Re: st: question about  Weighted Least Square(WLS)

this is discussed, briefly, on pp. 2305-2306 of [R] vwls

Rich

On 4/2/13 4:01 PM, Wu Zhang wrote:
>
> Hi folks,
>
>    I run two versions of WLS:
>
> 1.  regress y x [aweight = 1/exp(v)]
>
> 2. vwls y x , sd(sqrt(exp(v)))
>
> The estimates of parameters are exactly the same, while the standard errors are totally different: the ratio of the two standard errors for each estimate is a constant. Could someone explain how those result would happen?
>
> Thanks,
>
> Wu
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```