A characterization of moral transitive acyclic directed graph Markov models as labeled trees

被引:7
作者
Castelo, R [1 ]
Siebes, A [1 ]
机构
[1] Univ Utrecht, Dept Comp Sci, Inst Comp & Informat Sci, NL-3508 TB Utrecht, Netherlands
关键词
graphical Markov model; conditional independence; multivariate distribution; undirected graph; Markov model; acyclic directed graph Markov model; transitive acyclic directed graph Markov model; decomposable Markov model; lattice conditional independence Markov model; tree conditional independence Markov model; finite distributive lattice; poset; labeled tree;
D O I
10.1016/S0378-3758(02)00143-X
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
It follows from the known relationships among the different classes of graphical Markov models for conditional independence that the intersection of the classes of moral acyclic directed graph Markov models (or decomposable equivalent to DEC Markov models), and transitive acyclic directed graph equivalent to TDAG Markov models (or lattice conditional independence equivalent to LCI Markov models) is non-empty. This paper shows that the conditional independence models in the intersection can be characterized as labeled trees. This fact leads to the definition of a specific Markov property for labeled trees and therefore to the introduction of labeled trees as part of the family of graphical Markov models. (C) 2002 Elsevier Science B.V. All rights reserved.
引用
收藏
页码:235 / 259
页数:25
相关论文
共 25 条