[R] mulitmodal distributions

Bert Gunter gunter.berton at gene.com
Tue Nov 20 00:54:58 CET 2007


If you can write a density function then you can (usually easily in R) write
calculate a (log)likelihood that can be minimized by optim(). Of course, as
always, the data and starting values must support the parameterization.

I don't know anything about the paper you reference, but as a general
comment, methods that old were often devised to circumvent computational
issues that may today be irrelevant. Method of moments calculations often
fall into this class. 


Bert Gunter
Genentech Nonclinical Statistics


-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of Marion Wittmann
Sent: Monday, November 19, 2007 2:57 PM
To: r-help at r-project.org
Subject: [R] mulitmodal distributions

Hello, 
 
I see that "mclust" is a pacakge that handles fitting mixtures of normals.
Are there any other packages out that that can handle mixtures of gammas or
other exponentials? 
 
Additionally, are there any packages out there that can fit bimodal
distributions without mixtures? i.e., Cobb et al. 1983 using moment
recursion relations? 
 
Thank you! 
 
mw
 
Marion Wittmann, Ph.D. candidate
Environmental Science & Management
University of California
Santa Barbara, CA  93106-5131 
Lab: (805) 893 5890
Mobile: (805) 448 8259
Email:  <mailto:mwittmann at bren.ucsb.edu> mwittmann at bren.ucsb.edu
Web:  <http://www.bren.ucsb.edu/~mwittmann> www.bren.ucsb.edu/~mwittmann
 
 

	[[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org 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.



More information about the R-help mailing list