Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: VWLS for Trend Analysis


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: VWLS for Trend Analysis
Date   Thu, 11 Sep 2008 20:54:09 +0100 (BST)

--- jasonm@ucla.edu wrote:
> I have rates (=injuries/time) and their accompanying confidence  
> intervals for a 12 month period (ie 12 rates). I do not have the raw 
> data. I only have the point estimates and their accompanying CIs.
> Is -vwls- a good approach to see if a trend is statistically  
> significant (the point estimates have already been plotted)? Does  
> -vwls- have any kind of eform option for the coefficients (since I  
> took the log of all the variables)?

It seems more a situation for aweights rather than -vwls-, the variance
weighted least squares, deals with estimated standard deviations, while
what you got are standard errors. I wouldn't put too much faith in
those trends though. 

*--------------- begin example -------------
clear
input t b lb ub
1 2 1 3
2 3 2 4
3 3.5 1.5 4.5
4 4 3 5
end

gen se = (ub - lb)/(2* invnorm(.975))
gen w = 1/se^2
reg b t [aw = w]
*------------- end example ---------------
(For more on how to use examples I sent to the Statalist, see
http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )

-- Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      
*
*   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