Scholarlink

<p>This article introduces two generalizations of the spatial beta model that can effectively handle skewness and bimodality when analyzing bounded geostatistical data. The first approach is based on a mixture structure, while the second adopts the sinh-Gaussian random field to model spatial random effects. A key advantage of the second approach is its avoidance of using a mixture framework, which leads to fewer parameters and reduces the complexities associated with spatial mixture models. This makes the Bayesian inference implementation much easier. To evaluate the performance of the proposed models, we conducted simulation studies, and a comparison was made with the spatial beta model. The results from a practical example demonstrate that, in contrast to the mixture model, the second approach outperforms the competitor in spatial predictions.</p>

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

Generalized spatial beta models for skewed and bimodal proportion data

  • 期刊论文

  • 发表日期：

    2025年12月24日

查看全文
  • Mehdi Homayouni,
  • Majid Jafari Khaledi,
  • Esmaeil Najafi,
  • Hormoz Sohrabi

摘要

This article introduces two generalizations of the spatial beta model that can effectively handle skewness and bimodality when analyzing bounded geostatistical data. The first approach is based on a mixture structure, while the second adopts the sinh-Gaussian random field to model spatial random effects. A key advantage of the second approach is its avoidance of using a mixture framework, which leads to fewer parameters and reduces the complexities associated with spatial mixture models. This makes the Bayesian inference implementation much easier. To evaluate the performance of the proposed models, we conducted simulation studies, and a comparison was made with the spatial beta model. The results from a practical example demonstrate that, in contrast to the mixture model, the second approach outperforms the competitor in spatial predictions.