论文信息 - On the Existence of Maximum Likelihood Estimators for Graphical Gaussian Models

On the Existence of Maximum Likelihood Estimators for Graphical Gaussian Models

In graphical Gaussian models, it is well known that the maximum likelihood estimator exists with probability one if the number of replicates is at least as large as the number of variates. In this paper we deal with the case of fewer replicates. For the chordless p-cycle with two replicates, we prove that the maximum likelihood estimator exists with probability strictly between zero and one. (In case of p = 4 and independence, the probability is 2/3.)

Søren Ladegaard Buhl

[1] T. Speed,et al. On the Existence of Maximum Likelihood Estimators for Hierarchical Loglinear Models , 1988 .

[2] S. Lauritzen,et al. Mixed graphical association models; discussions and reply , 1989 .

[3] N. Wermuth. Analogies between Multiplicative Models in Contingency Tables and Covariance Selection , 1976 .

[4] J. N. R. Jeffers,et al. Graphical Models in Applied Multivariate Statistics. , 1990 .

[5] Vern I. Paulsen,et al. Schur Products and Matrix Completions , 1989 .

[6] T. Speed,et al. Gaussian Markov Distributions over Finite Graphs , 1986 .

[7] M. Frydenberg,et al. Decomposition of maximum likelihood in mixed graphical interaction models , 1989 .

[8] T. Speed,et al. Decomposable graphs and hypergraphs , 1984, Journal of the Australian Mathematical Society. Series A. Pure Mathematics and Statistics.