2016 Fiscal Year Final Research Report
Testing for multivariate normality
| Project/Area Number |
26330053
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Okayama University of Science (2016)
Kurashiki University of Science and the Arts (2014-2015) |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
NIKI NAOTO 東京理科大学, 工学部, 名誉教授 (10000209)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Keywords | 正規性 / 歪度 / 尖度 / キュムラント |
|---|
| Outline of Final Research Achievements |
As a first fruit, we give a new graphical method testing for normality based on an approximate joint probability density function of the sample skewness and kurtosis. In addition, we develop a new graphical method to access the multivariate normality combining with the principal component method.
Secondly, we give a recurrence formula for the joint distribution of the sample skewness and kurtosis. Moreover, we obtain a recurrence formula for their product moments and their exact expression up to the requisite order.
|
| Free Research Field |
統計科学
|
