I am looking for procedure that allow one to fit multiple distributions to a
variable. For example, based on analysis of the data we suppose that the data
can be represented by 3-5 normal distributions added together. I would like to
be able to determine the mean, sd, and weight associated with each distribution
and examine the improvement of the fit when 3,4 or 5 normal distributions are
used as components of the observed data.
The goal is to be able to separate out the background observations from impacted
observations and be able to develop summary stats that describe
background/baseline. I have been experimenting with mclust package but I am not
convinced this is the best/simplest way to proceed. I am hoping to add this sort
of analysis to some of the other ways were are able to characterized background
and compare them all. This is for the evaluation of chemicals in the
environment.
Michael J. Bock, PhD | Manager
ENVIRON | www.environcorp.com <http://www.environcorp.com/>
136 Commercial Street, Suite 402
Portland, ME 04101
V: 207.347.4413 x223| F: 207.347.4384 |mbock@environcorp.com
This message contains information that may be confidenti...{{dropped:11}}
