Decomposition of maximum likelihood in mixed graphical interaction models

M FRYDENBERG; Steffen L. Lauritzen

doi:10.1093/biomet/76.3.539

Decomposition of maximum likelihood in mixed graphical interaction models

50 Citations (Scopus)

Abstract

We show how decompositions of marked graphs induce corresponding decompositions of maximum likelihood estimates in mixed graphical interaction models. We use this result to obtain explicit formulae for the estimates in decomposable models and to obtain factorizations of likelihood ratio statistics.

Original language	English
Journal	Biometrika
Volume	76
Issue number	3
Pages (from-to)	539-555
Number of pages	17
ISSN	0006-3444
DOIs	https://doi.org/10.1093/biomet/76.3.539
Publication status	Published - 1989
Externally published	Yes

Access to Document

10.1093/biomet/76.3.539

Cite this

@article{c6a71c2e0ef84b4bb387346a02ac62f7,

title = "Decomposition of maximum likelihood in mixed graphical interaction models",

abstract = "We show how decompositions of marked graphs induce corresponding decompositions of maximum likelihood estimates in mixed graphical interaction models. We use this result to obtain explicit formulae for the estimates in decomposable models and to obtain factorizations of likelihood ratio statistics.",

author = "M FRYDENBERG and Lauritzen, {Steffen L.}",

year = "1989",

doi = "10.1093/biomet/76.3.539",

language = "English",

volume = "76",

pages = "539--555",

journal = "Biometrika",

issn = "0006-3444",

publisher = "Oxford University Press",

number = "3",

}

TY - JOUR

T1 - Decomposition of maximum likelihood in mixed graphical interaction models

AU - FRYDENBERG, M

AU - Lauritzen, Steffen L.

PY - 1989

Y1 - 1989

N2 - We show how decompositions of marked graphs induce corresponding decompositions of maximum likelihood estimates in mixed graphical interaction models. We use this result to obtain explicit formulae for the estimates in decomposable models and to obtain factorizations of likelihood ratio statistics.

AB - We show how decompositions of marked graphs induce corresponding decompositions of maximum likelihood estimates in mixed graphical interaction models. We use this result to obtain explicit formulae for the estimates in decomposable models and to obtain factorizations of likelihood ratio statistics.

U2 - 10.1093/biomet/76.3.539

DO - 10.1093/biomet/76.3.539

M3 - Journal article

SN - 0006-3444

VL - 76

SP - 539

EP - 555

JO - Biometrika

JF - Biometrika

IS - 3

ER -