---------- Forwarded message ----------
From: Steven Samuels <[email protected]>
Date: Apr 3, 2009 3:43 PM
Subject: Re: st:reweight to compare two samples
To: Mandy fu <[email protected]>
For the purposes of reweighting, you *should* specify that the new
weights are probability weights.
-Steve
On Apr 3, 2009, at 3:53 PM, Mandy fu wrote:
> Hi all,
>
> To compare wage rates of two samples, I 'll reweight sample 2 using
> sample 1 as the control group according to the gender and racial
> decomposition. The result of reweight will be the gender and racial
> decomposition of sample 2 is close to that in sample 1.
>
> But I'm not sure when I get the calculated "weight" from reweighting,
> as which type of weight I should treat it in following estimation in
> the four types of weights, like analysis weight, important weight,
> frequency weight(I know it should not be probability weight)?
>
>
*
* 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/
