2020 Fiscal Year Final Research Report
Statistical modeling for circular data in ecological and environmental science
Project/Area Number |
18K13459
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 12040:Applied mathematics and statistics-related
|
Research Institution | University of Shizuoka |
Principal Investigator |
Imoto Tomoaki 静岡県立大学, 経営情報学部, 助教 (20749296)
|
Project Period (FY) |
2018-04-01 – 2021-03-31
|
Keywords | 方向統計学 / 確率分布 / 数理統計学 |
Outline of Final Research Achievements |
In this research project, we constructed a probabilistic model to enable a flexible statistical analysis for data focusing on periodic phenomena such as angle and time. Specifically, we proposed a method for constructing circular distributions that can give flexible fitting to various circular data, a cylinder distribution that can model long-tailed for its linear part, and torus distributions that can be used with a low computational cost. Applications to seismic data and shared genes position data of the bacterial genomes show the tendency of big earthquakes and commonality of bacteria which has the same ancestor, respectively,
|
Free Research Field |
統計科学
|
Academic Significance and Societal Importance of the Research Achievements |
動物の移動方向や風向などの周期性のある現象に関するデータは、正規分布やポアソン分布のように直線上で定義される確率分布ではなく、円周上で定義される確率分布を用いて解析を行う必要がある。生態・環境科学分野ではこのようなデータが多くみられる。本研究で提案した柔軟かつ情報損失の少ない円周上分布、トーラス上分布、シリンダー上分布を用いることで、適切な生態動向推定や自然環境予測が可能となることが期待される。
|