Statistical mathematics for population dynamics models: information criterion, observation model, approximate Bayesian computation

2020 Fiscal Year Final Research Report

• Project/Area Number: 17K07578
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Ecology/Environment
• Research Institution: The Institute of Statistical Mathematics
• Principal Investigator: Shimatani Kenichiro 統計数理研究所, データ科学研究系, 准教授 (70332129)
• Project Period (FY): 2017-04-01 – 2021-03-31
• Keywords: 個体群動態 / 群集動態 / 状態空間モデル / ベイズ推定 / クローナル繁殖 / 標識調査 / 情報量規準

Outline of Final Research Achievements

Setting information criterion, observation model and approximated Bayesian computation as the three keywords, this study developed state space models in which a mathematical model is used for a state model and a statistical model is used for an observation model of field observations. For animals, for example, integrating capture-mark-recapture data under natural conditions and experimental results in a laboratory, two-species community dynamics was examined. For plants, integrating aboveground and belowground information into a state space model, population dynamics of a clonal plant was examined.

Free Research Field

生態統計学

Academic Significance and Societal Importance of the Research Achievements

異質な情報を一つの統計モデルに統合して定量的に分析する需要は、生態学に限らず様々な場面で強く求められている。本研究のような情報の統合の実例を積み上げることは、特定の分野を超えて統計手法の汎用性を高めていく効果をもたらす。