icf: Iterative conditional fitting

In ggm: Graphical Markov Models with Mixed Graphs

Iterative conditional fitting

Description

Main algorithm for MLE fitting of Gaussian Covariance Graphs and Gaussian Ancestral models.

Usage

icf(bi.graph, S, start = NULL, tol = 1e-06)
icfmag(mag, S, tol = 1e-06)

Arguments

bi.graph	a symmetric matrix with dimnames representing the adjacency matrix of an undirected graph.
mag	a square matrix representing the adjacency matrix of an ancestral graph (for example returned by makeAG).
S	a symmetric positive definite matrix, the sample covariance matrix. The order of the variables must be the same of the order of vertices in the adjacency matrix.
start	a symmetric matrix used as starting value of the algorithm. If start=NULL the starting value is a diagonal matrix.
tol	A small positive number indicating the tolerance used in convergence tests.

Details

These functions are not intended to be called directly by the user.

Value

Sigmahat	the fitted covariance matrix.
Bhat	matrix of the fitted regression coefficients associated to the directed edges.
Omegahat	matrix of the partial covariances of the residuals between regression equations.
iterations	the number of iterations.

Author(s)

Mathias Drton

References

Drton, M. & Richardson, T. S. (2003). A new algorithm for maximum likelihood estimation in Gaussian graphical models for marginal independence. Proceedings of the Ninetheen Conference on Uncertainty in Artificial Intelligence, 184–191.

Drton, M. & Richardson, T. S. (2004). Iterative Conditional Fitting for Gaussian Ancestral Graph Models. Proceedings of the 20th Conference on Uncertainty in Artificial Intelligence, Department of Statistics, 130–137.

See Also

fitCovGraph, fitAncestralGraph

ggm documentation built on May 29, 2024, 7:27 a.m.