|
Activity Number:
|
460
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #309021 |
|
Title:
|
The Evolutionary Forest
|
|
Author(s):
|
Scotland Leman*+
|
|
Companies:
|
Duke University
|
|
Address:
|
1800 Shelton Ave, Durham, NC, 27707,
|
|
Keywords:
|
Bayesian ; MCMC ; Phylogeny ; Population Biology
|
|
Abstract:
|
We describe a Markov chain Monte Carlo method for approximating the joint posterior distribution of parameters for evolutionary and other complex processes. Proposal distributions on complex structures such as phylogenies are essential for MCMC sampling methods. However, such proposal distributions are difficult to construct so that their probability distribution match that of the true target distribution, in turn hampering the efficiency of the overall MCMC scheme. We'll describe a data augmentation scheme that converges rapidly to the population parameters of interest, while utilizing a simple independent proposal distribution on individual trees. In this approach, we rely on an ensemble of histories (a forest of genealogical trees) rather than a single history. This enables the exploration of the augmented tree space (forest space) to proceed quickly and converge rapidly.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
