2016 Fiscal Year Final Research Report
The development of the multivariate nonparametric test for the environment data
Project/Area Number |
26730025
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Multi-year Fund |
Research Field |
Statistical science
|
Research Institution | Tokyo University of Science (2015-2016) 防衛大学校(総合教育学群、人文社会科学群、応用科学群、電気情報学群及びシステム工学群) (2014) |
Principal Investigator |
Murakami Hidetoshi 東京理科大学, 理学部第一部数理情報科学科, 講師 (60453677)
|
Project Period (FY) |
2014-04-01 – 2017-03-31
|
Keywords | ノンパラメトリック法 / 多変量データ / 仮説検定 |
Outline of Final Research Achievements |
In ecological statistics, the nonparametric statistical methods are often used to analyze the ecological data. We extended and proposed the nonparametric statistics for the multivariate data. In addition, we extended various statistics to the multiple comparison procedure in one-way layout in biometry. By a simulation study, the suggested statistics were more powerful than the other nonparametric statistics. Additionally, we derived the moment generating functions of statistics to approximate the exact distribution of the statistics. Moreover, we proved the unbiasedness of the test statistics. In data analysis, it is important to derive the distribution of the sum of independent non-identically gamma or uniform random variables. However, since the distributions include the sum of infinite series, it is difficult to apply for the real data analysis. Then, we derived the approximate distribution based on the saddlepoint approximation and the moment polynomial approximation.
|
Free Research Field |
統計科学
|