[S] More on xyplot in trellis with error bars

Frank E Harrell Jr (fharrell@virginia.edu)
Wed, 18 Mar 1998 12:20:18 -0500


Bill Venables pointed out to me that this problem was not fully specified.
Let me generalize it a bit. Suppose that one had a data frame with the
variables
month, treatment, pe, low, hi, where pe is the point estimate for some effect
and low
and hi are confidence limits corresponding to the point estimate (using
whatever approach). I want a line plot for pe vs. month with a separate panel
or
a separate line for each treatment, with bars connecting low to hi at the
value of month.
Venables & Ripley pg. 94-5 provide a line plot solution when (pe, low,hi)
constitute
different observations. In that case you can use groups= in xyplot to specify
what is plotted. What I would like trellis' xyplot to be able to do is accept
a matrix
on the left hand side of the formula (columns = pe low hi) but it won't. If
it accepted
a matrix I could pick the matrix apart inside a customized panel function.

---------------------------------------------------------------------------
Frank E Harrell Jr
Professor of Biostatistics and Statistics
Director, Division of Biostatistics and Epidemiology
Dept of Health Evaluation Sciences
University of Virginia School of Medicine
http://www.med.virginia.edu/medicine/clinical/hes/biostat.htm

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news