Effective inference and selection of statistical models to represent latent structure in spatial data
Project/Area Number |
26330042
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Statistical science
|
Research Institution | Okayama University |
Principal Investigator |
Sakamoto Wataru 岡山大学, 環境生命科学研究科, 教授 (70304029)
|
Project Period (FY) |
2014-04-01 – 2017-03-31
|
Project Status |
Completed (Fiscal Year 2016)
|
Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2016: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2015: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2014: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
|
Keywords | 階層 Bayes モデル / Markov 確率場 / モデル選択 / 情報量規準 / 領域同定 / 疾病地図データ / APC モデル / 周辺事後分布 / 階層ベイズモデル / マルコフ確率場 |
Outline of Final Research Achievements |
Some methods of making effective inference and selection in statistical models with high-dimensional latent variables are considered to reveal complicated latent structure in spacial and geographical data. It was shown that the method of detecting regions using estimated spatial effect, proposed for application to disease mapping data, had higher possibility of detecting regions with high risk than existing methods. Also it was suggested that the model selection method considered in the analysis of cancer mortality data with age-period-cohort (APC) models could estimate each effect appropriately, and give a new knowledge for interpretation.
|
Report
(4 results)
Research Products
(5 results)