Dear S+ Users:
I am attempting to understand a technique used in
climatology and meterology called the method of Partial Collectives. This
basically fits, based on some initial guesses, a series of normal
distributions to a multimodial empirical distribution (histogram). The
technique somehow minimizes the residual error and provides an assessment
of goodness of fit, allowing the means of the fitted distributions to be
interpreted as discontinuities in the empirical distribution.
Basically, I am wondering if anyone has experience with Partial
Collectives, and/or programs, or even better knows of an alternative but
equivalent technique implemented in S+.
best regards,
-------------------------------------------------------------------------------
Mike C. Sawada |Tel: 613-562-5800 x1734
Laboratory of Paleoclimatology and Climatology |Fax: 613-562-5145
Department of Geography |msawada@aix1.uottawa.ca
University of Ottawa
165 Waller Street
Ottawa, ON K1N 6N5
----------------- -----------------
************************************************************
WWW: http://aix1.uottawa.ca/academic/arts/geographie/lpcweb/
************************************************************
-------------------------------------------------------------------------------
-----------------------------------------------------------------------
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
|