Vines - A new graphical model for dependent random variables

被引:233
作者
Bedford, T
Cooke, RM
机构
[1] Univ Strathclyde, Dept Management Sci, Glasgow G1 1QE, Lanark, Scotland
[2] Delft Univ Technol, Fac Informat Technol & Syst, NL-2628 CD Delft, Netherlands
关键词
correlation; dependence; information; multivariate probability distribution; Monte Carlo simulation; tree dependence; Markov tree; belief net; multivariate normal distribution;
D O I
暂无
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 [统计学]; 070103 [概率论与数理统计]; 0714 [统计学];
摘要
A new graphical model, called a vine, for dependent random variables is introduced. Vines generalize the Markov trees often used in modelling high-dimensional distributions. They differ from Markov trees and Bayesian belief nets in that the concept of conditional independence is weakened to allow for various forms of conditional dependence. Vines can be used to specify multivariate distributions in a straightforward way by specifying various marginal distributions and the ways in which these marginals are to be coupled. Such distributions have applications in uncertainty analysis where the objective is to determine the sensitivity of a model output with respect to the uncertainty in unknown parameters. Expert information is frequently elicited to determine some quantitative characteristics of the distribution such as (rank) cot-relations, We show that it is simple to construct a minimum information vine distribution, given such expert information, Sampling from minimum information distributions with given marginals and (conditional) rank correlations specified on a vine can be performed almost as fast as independent sampling. A special case of the vine construction generalizes work of Joe and allows the construction of a multivariate normal distribution by specifying a set of partial correlations on which there are no restrictions except the obvious one that a correlation lies between -1 and 1.
引用
收藏
页码:1031 / 1068
页数:38
相关论文
共 25 条
[1]
[Anonymous], 1997, P ASA SECT BAYES STA
[2]
ARNOLD BC, 1999, CONDITIONAL SPECIFIC
[3]
COOKE RM, 1990, 1233 EST DELFT U TEC
[4]
COOKE RM, 1991, PROBABILISTIC SAFETY
[5]
COOKE RM, 1995, UNICORN METHODS CODE
[6]
PROBABILITY-DISTRIBUTIONS WITH GIVEN MULTIVARIATE MARGINALS AND GIVEN DEPENDENCE STRUCTURE [J].
CUADRAS, CM .
JOURNAL OF MULTIVARIATE ANALYSIS, 1992, 42 (01) :51-66
[7]
GENEST C, 1995, CR ACAD SCI I-MATH, V320, P723
[8]
GUIASU S, 1977, INFORMATION THEORY A
[9]
AN APPROACH TO SENSITIVITY ANALYSIS OF COMPUTER-MODELS .2. RANKING OF INPUT VARIABLES, RESPONSE-SURFACE VALIDATION, DISTRIBUTION EFFECT AND TECHNIQUE SYNOPSIS [J].
IMAN, RL ;
HELTON, JC ;
CAMPBELL, JE .
JOURNAL OF QUALITY TECHNOLOGY, 1981, 13 (04) :232-240
[10]
AN APPROACH TO SENSITIVITY ANALYSIS OF COMPUTER-MODELS .1. INTRODUCTION, INPUT VARIABLE SELECTION AND PRELIMINARY VARIABLE ASSESSMENT [J].
IMAN, RL ;
HELTON, JC ;
CAMPBELL, JE .
JOURNAL OF QUALITY TECHNOLOGY, 1981, 13 (03) :174-183