2013 Fiscal Year Final Research Report
Application of nonparametic methods in ecological statistics
| Project/Area Number |
23700349
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | 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) |
|---|
Principal Investigator |
MURAKAMI Hidetoshi 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, その他部局等, 講師 (60453677)
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Keywords | ノンパラメトリック法 / 検定統計量 / 多変量データ / 変化点問題 |
|---|
| Research Abstract |
In an environmental problem, the ecological statistics attracts attention in Statistics. In ecological statistics, the nonparametric statistical methods are often used to analyze the ecological data. However, by theoretical difficulty, the nonparametric methods for multivariate data had not been studied by many researchers. Therefore, we focused on the testing hypothesis for the univariate and multivariate two-sample problems and the detection of a change-point.
For a change-point problem, we proposed a nonparametric test and analyzed various real data. By a simulation study, the suggested statistic was more accurate than the other statistics. For univariate two-sample problem, we suggested the maxed-type nonparametric test and the modified Wilcoxon rank sum test. For multivariate two-sample problem, we considered various nonparametric rank sum tests based on the Jureckova-Kalina's ranks of distances. The proposed statistics were more powerful than the other nonparametric statistics.
|
