2008/02/29 by Hisayuki Hara, Akimichi Takemura
Computer Science · Environmental Science · Mathematics · #Algorithm #Artificial intelligence #Bayesian Modeling and Causal Inference #Computer science #Constructive #Data Management and Algorithms #Gaussian #Geometry #Graph #Graphical model #Iterative method #Mathematical optimization #Mathematics #Scaling #Soil Geostatistics and Mapping #Theoretical computer science #stat.CO #stat.ME
paper · pdf · doi:10.1080/03610920802238662
published as Communications in Statistics Theory and Methods, 39, No.8, 1643-1654, 2010 · 12 pages
arxiv created 2008/05/28 · openalex publication_date 2010/04/21 · arxiv updated 2010/07/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
We discuss an efficient implementation of the iterative proportional scaling procedure in the multivariate Gaussian graphical models. We show that the computational cost can be reduced by localization of the update procedure in each iterative step by using the structure of a decomposable model obtained by triangulation of the graph associated with the model. Some numerical experiments demonstrate the competitive performance of the proposed algorithm.