Expanding the regression analysis through the innovative applications of Bayesian methods
Project/Area Number |
20500259
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
Statistical science
|
Research Institution | Chuo University |
Principal Investigator |
YANAGIMOTO Takemi Chuo University, 理工学部, 客員教授 (40000195)
|
Project Period (FY) |
2008 – 2010
|
Project Status |
Completed (Fiscal Year 2010)
|
Budget Amount *help |
¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2010: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2009: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2008: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
Keywords | 交叉検証法 / 平滑化 / 不適切事前分布 / Bayes法 / DIC / GCV / 頻度法的接近 / hybrid接近 / e-混合 |
Research Abstract |
The assumption of a prior information for a parameter contained in the sampling density is essential, and the use of a proper prior density is obviously desired. This view allows us to pursue fundamental subjects such as the smoothing method. I began with exploring the deviance information criterion (DIC). Our tools for this challenging problem are the use of the e-mixture predictor and the two notions of the unbiasedness of a potential function. Then I attempted to explore a cross-validation criterion (CVC). The induced CVC fortunately shows good performance, and the reasons of performance becomes clear. The results are expected to stimulate our future studies of this subject.
|
Report
(4 results)
Research Products
(35 results)