Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Weighted average of historical observations |
Date | Fri, 27 Jan 2012 09:21:11 +0000 |
You could use -egen, total()- to get the individual sums, using -by:- or -by()- to do the calculation separately for individuals. Then divide. Nothing outside Stata required. Nick On Fri, Jan 27, 2012 at 4:52 AM, Samuel Yeo <yjs1987@gmail.com> wrote: > I have annual data on a variable, X, for a set of individuals. Given > their ages, I would like to generate a new variable X_HIS, which is a > weighted average of their past experiences of X through their > lifetimes. X_HIS would therefore incorporate all observations of X, > over each individual's lifetime. > > The calculation for X_HIS would be: > > WEIGHT FOR A GIVEN YEAR = (AGE - k)^m > where k is the time lag (in years), increasing from 1 to each > individual's AGE; m is an arbitrary parameter > > X_HIS = SUM(WEIGHT FOR A GIVEN YEAR * X) / (SUM OF WEIGHT FOR ALL YEARS) > > Is there a way to do this in Stata, or would I need to do it elsewhere > and import it in? * * 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/