Re: [s-news] S-News Daily Digest V2 #1954

[David L Lorenz <lorenz@usgs.gov>]

Doug,
  Is this what you want? You'll
need to recreate the rownames.
Dave
t(as.data.frame(
    lapply(
      lapply(DATA, function(x,
lag) acf(x,lag), lag=3), 
    function(x) x$acf[,,1])))

Doug Bolgiano <dougb@psbcresearch.org>
Sent by: s-news-owner@lists.biostat.wustl.edu

01/16/2009 11:51 AM

To	s-news@lists.biostat.wustl.edu
cc
Subject	Re: [S] [s-news] S-News Daily Digest V2 #1954

extracting components from a list of lists

I use tapply with the acf function (lag =3) and the subject index of a
data frame to calculate autocorrelations for a variable measured repeatedly
on each subject (n = 180).  The result is a list of lists (which are
the output of the acf function), one for each subject.  I want to
extract the $acf component from each subject's list and form a matrix where
rows index subjects and columns index lags.  

I have tried using the subscript operators [] and [[ ]] to extract the
acf componets.  Do I have to resort to looping through the subject
lists and extracting the acf component from each list individually or is
there a more efficient way to do this?

Doug Bolgiano
Puget Sound Blood Center
Seattle, WA  98104