Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: survival analysis


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: survival analysis
Date   Wed, 2 Sep 2009 21:10:17 +0000 (GMT)

--- On Wed, 2/9/09, Johannes Schoder wrote:
> Thanks for your help. However, when I am using your command Stata 
> complains that "weights must be stset".
> Therefore I used:
> stset time_in_months [weight=freq], failure(vital)
> 
> This works out. However I can't use aweight Stata automatically uses 
> fweight, if you should know how to actually tell Stata to use aweights I 
> would be very happy for any hints.

You can't use aweights with -stset-, as is implicit in the helpfile of 
-stset- (it is not in the list of allowed weight types). Anyhow, aweight
are virtually always the wrong weight type. It is there for the special 
case that your "data" are actually estimates, and the weights represent 
the precision of those estimates (sortof an early version of multi-level 
analysis). I think the weights should be something like the inverse of the 
standard error squared. The fweights are for the situation that a single 
row in your dataset represents multiple observations, so the values are 
not estimates but actual observed values and the weights say how much 
observations the row in your data represents. 

Hope this helps,
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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index