st: Re: solving equations with stata

["Michael Blasnik" <michael.blasnik@verizon.net>]

gen weight=res^-2
sum weight, meanonly
replace weight=weight*_N/r(sum)

Michael Blasnik
michael.blasnik@verizon.net


----- Original Message ----- 
From: "Gianfranco Spiteri" <Gianfranco.Spiteri@hpa.org.uk>
To: <statalist@hsphsun2.harvard.edu>
Sent: Wednesday, January 04, 2006 11:34 AM
Subject: st: solving equations with stata



> Hi,
> I am trying to perform a downweighting process for a dataset I am working 
> on
> (for an outbreak detection system of infectious diseases based on the
> Farrington method). As part of the process I have got to a point were I 
> have
> a list of residuals (res). I need to create weights for the original data
> based on these residuals using the following formulas:
> weight = a * res ^ -2
> where a is a constant such that:
> sum of all weights = n
>
> where n is the number of observations (n is 35 or less with this method)
> I'm sure there is a relatively straight forward way of doing this in 
> stata,
> but I cant seem to find it. Can anyone help me out,
> thanks
> Gianfranco
*
*   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/