is.Gident {ggm}R Documentation

G-identifiability of an UG

Description

Tests if an undirected graph is G-identifiable.

Usage

is.Gident(gmat)

Arguments

gmat a symmetric matrix with dimnames representing the edge matrix of an undirected graph

Details

An undirected graph is G-identifiable if every connected component of the complementary graph contains an odd cycle (cfr. Stanghellini and Wermuth, 2003). The algorithm used follows suggestions by Tarantola and Vicard (2002).

Value

a logical value, TRUE if the graph is G-identifiable and FALSE if it is not.

Author(s)

Giovanni M. Marchetti

References

Stanghellini, E. & Wermuth, N. (2003). On the identification of directed acyclic graph models with one hidden variable. Submitted and available at http://psystat.sowi.uni-mainz.de.

Stanghellini, E. (1997). Identification of a single-factor model using graphical Gaussian rules. Biometrika, 87, 199–205.

Tarantola, C. & Vicard, P. (2002). Spanning trees and identifiability of a single-factor model. Statistical Methods & Applications, 11, 139–152.

Vicard, P. (2000). On the identification of a single-factor model with correlated residuals. Biometrika, 87, 199–205.

See Also

UG, cycleMatrix

Examples

## A not G-identifiable UG
G1 <- UG(~ a*b + u*v)
is.Gident(G1)
## G-identifiable UG
G2 <- UG(~ a + b + u*v)
is.Gident(G2)

[Package Contents]