Research on variable selection for high-dimensional varying coefficient models
| Project/Area Number |
25400197
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Foundations of mathematics/Applied mathematics
|
|---|
| Research Institution | Hitotsubashi University |
|---|
Principal Investigator |
HONDA Toshio 一橋大学, 大学院経済学研究科, 教授 (30261754)
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2014: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2013: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 変数選択 / 高次元データ / 変動係数モデル / セミパラメトリックモデル / longitudinal data |
|---|
| Outline of Final Research Achievements |
I considered variable selection for high-dimensional varying coefficient models for longitudinal data and obtained some significant results. Specifically, I proposed with collaborators a useful two-step procedure to identify the true semi-varying coefficient model of the longitudinal data. The procedure consists of the nonparametric independence screening and the group SCAD method. Besides, I proposed an efficient estimator of the constant coefficients of the true semi-varying coefficient. I proved its desirable theoretical properties and some numerical properties were also investigated in our collaborative papers. In addition, I also obtained some theoretical results for a forward variable selection procedure for high-dimensional variable coefficient models and Cox models with variable coefficients. Numerical properties were also examined in our collaborative papers.
|
Report
(4 results)
Research Products
(16 results)
