| Project/Area Number |
16K03590
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | Otaru University of Commerce |
|---|
Principal Investigator |
LIU QINGFENG 小樽商科大学, 商学部, 教授 (60378958)
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2018: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2017: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
|
|---|
| Keywords | Model Selection / Model Averaging / Factor / Nonlinear / Optimality / High-Dimensional / Algorithm / Sparsity / Model selection / Model averaging / Volatility model / Diffusion index model / Factor model / 非線形回帰モデル / GARCH型モデル / モデル選択 / モデル平均 / 情報量基準 / 最適性 / 非線形モデル / GARCHモデル / ファクター抽出 / Diffusion Index model / Factor Selection / Big Data / Forecasting |
|---|
| Outline of Final Research Achievements |
5 research papers have been written. 2 of them have been published. The other 3 have been reported in some international conferences or seminars. The most important paper considered the problem of model averaging estimation for regression models that can be nonlinear in their parameters and variables. We proved the optimality of the new method. Monte Carlo experiments revealed that NMA lead to relatively lower risks compared with alternative model selection and model averaging methods in most situations. Empirical results showed that in most cases, our method leads to the lowest prediction errors. Moreover, I and my coauthors proposed model averaging methods of GARCH type models for analyzing financial data in three papers, which contribute to forecasting in financial market. In the last year, I showed the sparsity of the weight of the model averaging method for linear models. Subsequently, a high-speed algorithm for the model averaging method was proposed.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究は経済現象や他の社会現象や自然現象の分析のために新しい統計的方法を開発した。特にビッグデータの一種である高次元データの分析の精度を高めることやその分析のための計算コストを下げることに貢献している。統計理論を発展させると同時に、実用的な研究成果であるため、様々な分野で応用されて国民経済の発展に貢献できると期待する。
|
