| Project/Area Number |
15K03405
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | Ryukoku University (2018)
Setsunan University (2015-2017) |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
作道 真理 一般財団法人日本経済研究所, 調査局, 研究員(移行) (70748954)
|
|---|
| Project Period (FY) |
2015-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2018: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2017: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2015: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
|
|---|
| Keywords | ノンパラメトリック計量経済学 / 接合データの計量経済学 / 非対称カーネル関数 / 回帰不連続デザイン |
|---|
| Outline of Final Research Achievements |
Nonparametric and semiparametric methods in which the properties peculiar to economic data sets are taken into account are improved and developed. The properties are as follows. First, many economic variables take only nonnegative values by definition. In addition, their distributions have many observations concentrated in the vicinity of the origin, whereas they also possess a long tail with sparse data. Second, it is often the case that a single data set does not contain all necessary variables for regression analyses in empirical microeconomics. For the former, asymmetric kernel functions are applied to a variety of estimation and testing problems. For the latter, methods of estimating regression models consistently in which the missing regressors are imputed nonparametrically from another sample are explored.
|
| Academic Significance and Societal Importance of the Research Achievements |
ノンパラメトリック推定・検定問題に対する非対称カーネル関数の利用可能性を拡大した。また、経済・ファイナンスにおける非対称カーネル関数の応用を取り扱う初の書籍も出版した。さらに、以前から実務家の間では二標本を接合したデータを用いて最小二乗推定を行う慣行が存在したが、一致性の確保、収束速度等統計的に重要な問題は見落とされていた。本研究ではこれらの問題に対する回答を与えた。
|
