| Project/Area Number |
18K11199
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 60030:Statistical science-related
|
|---|
| Research Institution | Tokyo University of Science |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2023-03-31
|
| Project Status |
Completed (Fiscal Year 2022)
|
| Budget Amount *help |
¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Fiscal Year 2020: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2019: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2018: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | ノンパラメトリック法 / Ranked Set Sampling / 近似分布 / 密度推定 / 検定統計量 / カーネル密度推定 |
|---|
| Outline of Final Research Achievements |
Ranked set sampling (RSS) is one of random sampling methods from population. RSS is widely used in many scientific fields. In this research, we proposed one-, two- and multisample nonparametric test statistics in testing problem. We derived the limiting distributions and the approximate distributions of proposed test statistics. In addition, we showed theoretical properties such as the asymptotic power, the consistency and the unbiasedness. We also suggested how to determine the rank of vector value of data. Since, in practice, the distribution of RSS data is unknown, we used the kernel density estimation to estimate the population distribution and parameters.
|
| Academic Significance and Societal Importance of the Research Achievements |
一般的な仮説検定問題では，単純無作為抽出から得られたデータに対して分析を行なう．しかし，多種多様なデータが存在する現代社会においては，様々なサンプリング方法によって得られたデータが存在するため，既存の手法では対処できない問題が多々発生する．本研究成果は，その解決方法の１つとして，統計学のさらなる発展を担うものである．
|
