1997 Fiscal Year Final Research Report Summary
DEVELOPMENT OF ANALYSIS SYSTEM FOR QUALITY INFORMATION BY GRAPHICAL MODELLING
| Project/Area Number |
08308024
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
社会システム工学
|
|---|
| Research Institution | TOKYO INSTITUTE OF TECHNOLOGY (1997)
The University of Tokyo (1996) |
|---|
Principal Investigator |
MIYAKAWA Masami Tokyo Institute of Technology, Graduate School of Decision Science and Technology, Associate Professor, 大学院・社会理学工研究科・経営工学専攻, 助教授 (90157595)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NISHINA Ken Nagoya Institute of Technology, Department of Engineering, Associate Professor, 工学部, 助教授 (60115681)
NAGATA Yasushi Okayama University, Department of Economics, Associate Professor, 経済学部, 助教授 (30198337)
TSUBAKI Hiroe Tsukuba University, Graduate School of Policy and Planning Sciences, Associate P, 大学院・社会工学系, 助教授 (30155436)
KURIKI Satoshi The Institute of Statistical Mathematics, Associate Professor, 統計基礎研究系, 助教授 (90195545)
IIDA Takahisa Keio University, Department of Science and Engineering, Research Assintant, 理工学部, 助手 (00114851)
|
|---|
| Project Period (FY) |
1996 – 1997
|
| Keywords | Statistical causal analysis / Covariance selection model / Log-linear model |
|---|
| Research Abstract |
A graphical model is a probability model for random observations whose independence structure can be characterised by independence graph, and graphical modelling is the statistical activity of fitting graphical models to the data. The theory of graphical modelling has emerged from a mixture of log-linear models in multi-way contingency tables and covariance selection models in multivariate normal distribution. In this research, we develop the conversational data analysis system both for graphical Gaussian modelling which can be used for analysis of multi-dimensional continuous variables when assuming their distribution to be multivariate normal distribution, and for graphical log-linear modelling which can be used for multi-dimensional discrete variables when assuming their distribution to be mulinomial distribution. This system is applicable to representation of a probability model by using not only undirected independence graphs but also directed and chain independence graphs. The system can be linked to the existing analysis system for other methods of multivariate analysis.
|
Research Products
(12 results)
