Fwd: Re: st: how to weight (level-1) data using gllamm

[kasakawa@ualberta.ca]

Dear Chloe

when using weight( ) option in gllamm, if you want to refer to level-1  
weight, you cannot name your weight variable as simply "wt".  If your  
weight refers to level-1 weight, you will have to name your weight  
variable as "wt1" in your dataset. If weight is at level 2, then you  
will have to name it as "wt2", then use option

weight(wt)

Then if your dataset contains wt1 then STATA knows that your  
weight(wt) refers to your varibale called wt1.

See page 22 of gllamm manual

http://www.bepress.com/cgi/viewcontent.cgi?article=1160&context=ucbbiostat

for weight( ) option

KA


----- Forwarded message from kasakawa@ualberta.ca -----
    Date: Wed, 02 May 2007 11:57:06 -0600
    From: kasakawa@ualberta.ca
Reply-To: kasakawa@ualberta.ca
 Subject: Re: st: how to weight (level-1) data using gllamm
      To: statalist@hsphsun2.harvard.edu

If your weight represents survey weights (i.e. probabilty weights),
then you will have to use pweight(wt) option and call your weight
variable as "wt1" not "wt" (I think gllamm recognizes "wt1" as level-1
weight).   Or, sometimes Stata variable is case-sensitive, so check
your variable name whether or not it is called "wt" or "WT".  If the
variable name is in an upper-case letters, then you will have to use
upper-case letters in your syntax as well.

Regards,
KA

Quoting Tsui-o Tai <tsuiot@yahoo.com>:


> Hi Stata experts,
>
> I got problems when I used "gllamm" and tried to
> weight (level-1 weight). My syntax is like:
> gllamm poor d1 sex marry elder d27..., i (country) nrf
> (1) eqs (inter) link (logit) family (binom) weight
> (wt);
>
> In the output, wt is not recognized even though I
> generated wt already. Please a me an syntax example
> and tell me how to weight data using gllamm. Your help
> is deeply appreciated.
>
> Chloe
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/



----- End forwarded message -----


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/