Bayesian parameter inference for individual-based models using a Particle Markov Chain Monte Carlo method

详细信息    查看全文

• 作者：Mira Kattwinkel ; ¹ ; ^{mira.kattwinkel@gmx.net} ; Peter Reichert
• 关键词：Parameter estimation ; Calibration ; Agent-based model (ABM) ; Individual-based model (IBM) ; Bayesian inference ; Particle Markov Chain Monte Carlo (PMCMC) ; Approximate Bayesian Computation (ABC)
• 刊名：Environmental Modelling & Software
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：87
• 期：Complete
• 页码：110-119
• 全文大小：1104 K
• 卷排序：87

文摘

We present a new option for Bayesian inference for agent-/individual-based models. The implementation is based on Particle Markov Chain Monte Carlo (PMCMC). We found good performance and insights into model behaviour for a simple IBM with artificial data. Bayesian inference for AMBs/IBMs stimulates mechanistic ecological research.