A study on theory and applications of Bayesian estimation for parameter matrices in multivariate statistical models
| Project/Area Number |
23500361
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Toho University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Project Status |
Completed (Fiscal Year 2013)
|
| Budget Amount *help |
¥2,600,000 (Direct Cost: ¥2,000,000、Indirect Cost: ¥600,000)
Fiscal Year 2013: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2012: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2011: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 統計的推測 / ベイズ推定 / 統計的決定理論 / 縮小推定 / スタイン現象 / 共分散行列 / 多変量推測理論 / 縮小型推定 / ミニマクス推定量 / ミニマクス推定 |
|---|
| Research Abstract |
This study addresses the Bayesian estimation problems of parameter matrices in multivariate statistical models which are used for analyses of two dimensional array data. The principal aim of the study is to develop optimal Bayes estimators and shrinkage-type estimators not only from the point of view of statistical decision theory, but also from that of practical application. Some results are obtained for a least favorable sequence of prior distributions in covariance estimation, dominance properties of a generalized Bayes estimator of a bounded precision matrix, inadmissibility of shrinkage estimators for a mean matrix, and others.
|
Report
(4 results)
Research Products
(13 results)
