A flexible model called the max-mixture model has been introduced for modeling situations where the extremal dependence structure type may vary with distance. In this paper, we propose a novel estimation procedure for spatial max-mixture model parameters. Our procedure is based on the madogram, a dependence measure used in geostatistics to describe spatial structures. A nonlinear least squares minimization procedure is applied to obtain the estimators for extremal dependence functions. A simulation study shows that the proposed procedure works well for these models. In an analysis of monthly maxima of daily rainfall data collected over the East of Australia, we implement the proposed estimation procedure for diagnostic and confirmatory purposes.

中文翻译：

---

拟合具有未知极值依赖类的空间最大混合过程：探索性分析工具

引入了称为最大混合模型的灵活模型，用于建模极端依赖结构类型可能随距离变化的情况。在本文中，我们提出了一种针对空间最大混合模型参数的新颖估计程序。我们的程序基于madogram，这是在地统计学中用于描述空间结构的依赖度量。应用非线性最小二乘最小化程序来获得极值依赖函数的估计量。仿真研究表明，所提出的程序对于这些模型非常有效。在对澳大利亚东部地区收集的每日降雨量数据的每月最大值进行分析时，我们将提出的估算程序用于诊断和确认目的。