Skip to content

Randomly Stopped Sum Model

2 messages · Glen,George, Mollie Brooks

#
Hi all,

I found this paper online (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4226152), which discusses a method to model the sum of continuous random variables over a number of events where the number of events is also a random variable. I was wondering if anyone knew of specific examples using this approach to model reproductive events in animal populations? Or if a robust modeling framework in R has been constructed?

?What is best in mathematics deserves not merely to be learnt as a task, but to be assimilated as a part of daily thought, and brought again and again before the mind with ever-renewed encouragement.? Bertrand Russell

Graduate Assistant (PhD)
Archie Carr Center for Sea Turtle Research
Department of Biology
University of Florida

Tel: (+1) 352-282-5616         Website: http://accstr.ufl.edu/
Skype: geoglen12                  Email: george.glen at ufl.edu<mailto:george.glen at ufl.edu>
#
Hi George, 

I?m not familiar with the method in that paper, but I know that "a Tweedie distribution is equivalent to the distribution obtained by summing a Poisson number of gamma random variables" (Foster & Bravington 2013) and there are a few methods available in R for estimating the Tweedie distribution (i.e. compound Poisson?gamma distribution). 

?but is the measurement for each reproductive event really continuous, e.g something like total clutch weight or volume?

cheers,
Mollie