[R] Confidence intervals on Lowess (eg SiZer from J.S.Marron)

From: Dalphin, Mark <mdalphin_at_amgen.com>
Date: Fri 08 Sep 2006 - 19:38:06 GMT


I have some very noisy, relatively sparse data; a biological response of roughly
~8 subjects at ~8 times points). I've been following the data trend using a lowess
line, over-plotted with several values of bandwidth, 'f <- seq(0.3, 0.9, by=0.1)'.
At this point, we have no models for these data.

I wonder if there is any way under R to assign some sort of confidence interval to the
lowess line. For example, I have seen a method from the lab of J.S.Marron, which
his group has implemented in the Matlab program, SiZer.

        http://www.stat.unc.edu/faculty/marron/DataAnalyses/SiZer_Intro.html and, the more interesting:         


An implementation under R of SiZer would obviously answer this question; does it exist?
Suggestions of alternative approaches would also be welcome.

My searchs of the R-maillist archive for 'sizer', 'Marron' and 'scale-space' didn't return anything
that I recognized as useful here, but I am new to the idea of confidence in a lowess line
so I may not be using the appropriate vocabulary.


Mark Dalphin
Dept Comp Biol, M/S AW2/D3262
Amgen, Inc.
1201 Amgen Court W
Seattle, WA 98119
Phone: +1-206-265-7951

R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Sat Sep 09 05:43:13 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Fri 08 Sep 2006 - 20:17:45 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.