Research Repository

Perfect sampling methods for random forests

Dai, Hongsheng (2008) 'Perfect sampling methods for random forests.' Advances in Applied Probability, 40 (3). pp. 897-917. ISSN 0001-8678

Dai_revised.pdf - Accepted Version

Download (319kB) | Preview


<jats:p>A weighted graph<jats:italic>G</jats:italic>is a pair (<jats:italic>V</jats:italic>, ℰ) containing vertex set<jats:italic>V</jats:italic>and edge set ℰ, where each edge<jats:italic>e</jats:italic>∈ ℰ is associated with a weight<jats:italic>W<jats:sub>e</jats:sub></jats:italic>. A subgraph of<jats:italic>G</jats:italic>is a forest if it has no cycles. All forests on the graph<jats:italic>G</jats:italic>form a probability space, where the probability of each forest is proportional to the product of the weights of its edges. This paper aims to simulate forests exactly from the target distribution. Methods based on coupling from the past (CFTP) and rejection sampling are presented. Comparisons of these methods are given theoretically and via simulation.</jats:p>

Item Type: Article
Uncontrolled Keywords: Coupling from the past; MCMC; perfect sampling; rejection sampling; trees and forests
Subjects: Q Science > QA Mathematics
Divisions: Faculty of Science and Health
Faculty of Science and Health > Mathematical Sciences, Department of
SWORD Depositor: Elements
Depositing User: Elements
Date Deposited: 12 Feb 2013 08:07
Last Modified: 15 Jan 2022 00:54

Actions (login required)

View Item View Item