2006 Fiscal Year Final Research Report Summary
Analysis for contingency tables by means of algebraic technique
| Project/Area Number |
17500187
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Kurashiki University of Science and the Arts |
|---|
Principal Investigator |
NAKAGAWA Shigekazu Kurashiki University of Science and the Arts, Dept.of Computer Science and Mathematics, Professor, 産業科学技術学部, 教授 (90248203)
|
|---|
| Project Period (FY) |
2005 – 2006
|
| Keywords | mathematical statistics / applied mathematics / algorithm / foundation of information science |
|---|
| Research Abstract |
The analysis with the algebraic technique which includes in Groebner basis theory as one of the recent research direction is observed in statistics, especially in contingency table analysis.
Presently, we can implement MCMC (Markov Chain Monte Carlo) on the contingency table on the computer programs.
It is designed in such a way that it can execute the p value calculation in the contingency table. Those where it is done until now are in 2 way contingency tables and no interaction model on 3 way contingency tables.
In this research, we have calculated p value for the contingency table which has the structure of especially certain kind, namely, the conditional independent model in 4 way contingency tables. Our algorithm is implemented not only on R language but also on C language. Our algorithm suggests us to apply to the graphical model on the multi way contingency tables.
Furthermore, during particular research period, we have investigated on the distribution of the sample Pearson measure of skewness and test for normality.
|
