Marginalizing and conditioning in graphical models (Q1860998)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Marginalizing and conditioning in graphical models
scientific article

    Statements

    Marginalizing and conditioning in graphical models (English)
    0 references
    0 references
    0 references
    25 May 2003
    0 references
    The author introduces the class of MC graphs and shows that it is closed under the operations of marginalizing over a certain subset of vertices, or conditioning on a subset of other vertices. The present paper addresses the marginalization and conditioning problem for a large class of graphs, containing at least all undirected graphs and all directional graphs. Directional graphs are important, since they occur as path diagrams in Gaussian linear structural equation systems.
    0 references
    0 references
    0 references
    0 references
    0 references
    Gaussian linear equation systems
    0 references
    graph
    0 references
    graphical Markov model
    0 references
    linear structural equation systems
    0 references
    Markov property
    0 references
    Markov random field
    0 references
    MC graph
    0 references
    non-recursive casual model
    0 references
    separation
    0 references