2011 Fiscal Year Final Research Report
Selecting an optimal mixed effect model based on simulation : theory and applications
| Project/Area Number |
21500275
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
SAKAMOTO Wataru 大阪大学, 大学院・基礎工学研究科, 准教授 (70304029)
|
|---|
| Project Period (FY) |
2009 – 2011
|
| Keywords | モデル選択 / 情報量規準 / 制限付き最尤推定 / モンテカルロ法 / パラメータ制約モデル |
|---|
| Research Abstract |
Mixed effect models are useful for analysis of data which are measured repeatedly for each individual, and have been applied to various fields such as medical and biological sciences. However, how to select an optimal mixed effect model has not been fully examined. We proposed a criterion to select such an optimal model, which would be appropriate in theory and practically more useful. The criterion can be computed approximately by using random numbers. We confirmed that the proposed criterion should lessen risk of choosing a parsimonious model wrongly.
|
