Sorry. I should have clarified. The reason I am using aweights are
similar to what you said. My observations are not really observations
but more of estimates.
I figured an easier way around. Pls let me know what you think:
Step 1. ttest age1 == age2 [aweight = wt], unpaired
Step 2 lincom (age1)-(age2) to get the t-test and probabilities
thanks,
Sripal.
On Tue, Nov 10, 2009 at 10:04 AM, Maarten buis <[email protected]> wrote:
> --- Sripal Kumar wrote:
>> I am trying to use t-test of independent samples but by
>> using analytic weights. I tried the following:
>>
>> ttest age1 == age2 [aweight = wt], unpaired
>>
>> I get an error message saying "weights not allowed".
>> Is there a way around it?
>
> --- On Tue, 10/11/09, Martin Weiss wrote:
>> The archive is full of advice to try an equivalent
>> -regress- command...
>
> The archive is also full of advise that aweights are almost
> never the appropriate weight type (aweights are for the
> very special situation where your observations aren't
> observations but estimates and the weights are the inverse
> of the standard error squared).
>
> -- Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://www.maartenbuis.nl
> --------------------------
>
>
>
>
> *
> * 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/
