Is there a ready means to do mixture models of binned data in R?
I have a two dimensional grid of intensity values that represents many individual observations that are known to have bivariate normal distributions, although there is frequent overlap between the distributions. There is also a degree of noise in the data that may be modeled by a poisson distribution. I am looking for a way to identify the mixture model that best describes a subset of this grid of intensities so that I can deconvolute the multiple observations and noise distributions for quantification purposes.
I thought Mclust might be of use, but I didn't see any support for binned data. I have searched the archives, read the FAQ, and the document "Fitting Distributions with R" without finding a successful solution.
Any help would be appreciated.
Cheers, Leif Rustvold
______________________________________________ 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