| Project/Area Number |
19300088
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Cognitive science
|
|---|
| Research Institution | Chiba University |
|---|
Principal Investigator |
WANG Jinfang Chiba University, 大学院・理学研究科, 准教授 (10270414)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
HARA Hisayuki 東京大学, 大学院・情報理工学系研究科, 助教 (40312988)
TSUKUMA Hisayuki 東邦大学, 医学部, 講師 (50424685)
兪 文偉 千葉大学, 大学院・工学研究科, 准教授 (20312390)
田栗 正章 千葉大学, 大学入試センター, 副所長 (10009607)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
TAGURI Masaaki 千葉大学, 大学入試センター, 副所長 (10009607)
YU Wenwei 千葉大学, 大学院・工学研究科, 教授 (20312390)
|
|---|
| Project Period (FY) |
2007 – 2009
|
| Project Status |
Completed (Fiscal Year 2009)
|
| Budget Amount *help |
¥10,010,000 (Direct Cost: ¥7,700,000、Indirect Cost: ¥2,310,000)
Fiscal Year 2009: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Fiscal Year 2008: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
Fiscal Year 2007: ¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
|
|---|
| Keywords | cain algebra / cain polynomial / conditional independence / contingency table / graphical model / Stein's phenomenon / Markov base / shrinkage estimator / generalized Bayes estimator / logistic regression / matrix normal distribution / Admissibility / Bayes estimaton / Cain polynomial / Conditional independence / Graphical model / Normal mean / Statistical matching / bootstrap / graphical models / Kullback-Leibler loss / longitudinal data / maximum likelihood estimator / Shrinkage estimator |
|---|
| Research Abstract |
Probabilistic conditional independence (PCI) is a key for building statistical causal models, and we established an algebraic framework called the cain algebra for dealing with PCI. It is shown that the cain algebra is consistent with both the graphoid and the separoid. Further, we defined a new type of polynomials called the cain polynomials and showed that statistical models defined in terms of cain polynomials may be derived using tools of the cain polynomials.
|
