R help - May 2001 - Using "R" to find peaks in sampled data

This is probably a trivial question, but I did
not see it in the FAQs.  Perhaps I simply need
to be pointed to some URLs.

In any case, I have some sampled data that has peaks,
and I would like to extract peak height, location, and
width, according to almost any statistical model.
Gaussian would seem to make the most sense.

There can also be "steps" increasing the amplitude
over a region, that should also be captured and removed
before the peak analysis.

Can anyone point me to either some FAQs, or "R" code
to do this?  I have not yet installed "R", and this
is the only current use I would have for it and I want
to know if it makes sense to bother before going to
the effort.

TIA

Karl Sirotkin
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

> 
> In any case, I have some sampled data that has peaks,
> and I would like to extract peak height, location, and
> width, according to almost any statistical model.
> Gaussian would seem to make the most sense.
Be specific about the form of your data.
Do you mean that you have a single variable you measured, and you are
talking about peaks of some histogram of the data? Or do you have
x,y data? If the latter, the peak height is just max(y), and the location
is x[y==max(y)]. The width is not too hard; I have a function that finds
the
FWHM.

Bill Simpson

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

>>>>> "Bill" == Bill Simpson <wsi at gcal.ac.uk>
writes:
    >>  In any case, I have some sampled data that has peaks, and I would
    >> like to extract peak height, location, and width, according to
    >> almost any statistical model.  Gaussian would seem to make the most
    >> sense.
    Bill> Be specific about the form of your data.  Do you mean that you
    Bill> have a single variable you measured, and you are talking about
    Bill> peaks of some histogram of the data? Or do you have x,y data? If
    Bill> the latter, the peak height is just max(y), and the location is
    Bill> x[y==max(y)].

where you might want to use

      imax <- which.max(y)
      x[imax]
etc

    Bill>  The width is not too hard; I have a function that finds the FWHM.

which reminds me that S-plus has a function  peaks() 
that some folks have wanted for R.  If anyone has good code,
not stolen from S-plus, but still compatible,  please post...

Martin Maechler <maechler at stat.math.ethz.ch>
http://stat.ethz.ch/~maechler/
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._