Statistical Science

Linear Dependencies Represented by Chain Graphs

D. R. Cox and Nanny Wermuth

Full-text: Open access

Abstract

Various special linear structures connected with covariance matrices are reviewed and graphical methods for their representation introduced, involving in particular two different kinds of edge between the nodes representing the component variables. The distinction between decomposable and nondecomposable structures is emphasized. Empirical examples are described for the main possibilities with four component variables.

Article information

Source
Statist. Sci., Volume 8, Number 3 (1993), 204-218.

Dates
First available in Project Euclid: 19 April 2007

Permanent link to this document
https://projecteuclid.org/euclid.ss/1177010887

Digital Object Identifier
doi:10.1214/ss/1177010887

Mathematical Reviews number (MathSciNet)
MR1243593

Zentralblatt MATH identifier
0955.62593

JSTOR
links.jstor.org

Keywords
Chain model conditional independence covariance selection decomposable model linear structural equation multivariate analysis path analysis

Citation

Cox, D. R.; Wermuth, Nanny. Linear Dependencies Represented by Chain Graphs. Statist. Sci. 8 (1993), no. 3, 204--218. doi:10.1214/ss/1177010887. https://projecteuclid.org/euclid.ss/1177010887


Export citation