This study introduces a robust alternative to traditional species distribution models (SDMs) using Poisson Point Processes (PPP) and new divergence measures. We propose the F-estimator, a method grounded in cumulative distribution functions, offering enhanced accuracy and robustness over maximum likelihood (ML) estimation, especially under model misspecification. Our simulations highlight its superior performance and practical applicability in ecological studies, marking a significant step forward in ecological modeling for biodiversity conservation.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Minimum Divergence for Poisson Point Process

  • Shinto Eguchi

摘要

This study introduces a robust alternative to traditional species distribution models (SDMs) using Poisson Point Processes (PPP) and new divergence measures. We propose the F-estimator, a method grounded in cumulative distribution functions, offering enhanced accuracy and robustness over maximum likelihood (ML) estimation, especially under model misspecification. Our simulations highlight its superior performance and practical applicability in ecological studies, marking a significant step forward in ecological modeling for biodiversity conservation.