Generalized Linear and Nonlinear Mixed Models for Longitudinal and Spatial Data

物种在时间和空间上的丰富度在当代生态学，保护生物学和自然资源管理中具有极大的兴趣。在竞争性群落中，在同一地点观察到的不同物种的植物或动物的丰度通常是负相关的，而由于共享的环境和其他潜在的群落特征，同一物种的丰度往往是正相关的。由于杂草群落既受肥料的直接影响，又受作物竞争效应的间接影响，因此这种相关性既受共同的环境特征影响，也受竞争影响。我的研究兴趣在于多变量随机效应模型的发展，其中随机效应可以用来解释社区效应和相邻社区之间的关联。我们新的统计方法可以帮助寻找物种的群落组成模式及其与现有环境特征的关系。例如，预测野生动物种群的空间分布是制定保护野生动物管理战略的一个重要组成部分。了解物种在时间和空间上的丰富程度有助于决策者在适当的时候在地理上确定资源分配的目标，以试图补救问题地区。

The abundance of species over time and space is of great interest in the contemporary ecology, conservation biology, and natural resource management. Observed abundance of different species of plants or animals at the same location is usually negatively correlated in a competitive community, whereas abundance of the same specie tends to be positively correlated because of shared environment and other underlying community characteristics. As the weed community is influenced both directly by fertilizers and indirectly through the effect of crop competition, the correlation is influenced by both shared environmental characteristics and competition. My research interests lie in the development of multivariate random effects models where random effects can be used to account for community effects and the association among neighbouring communities. Our new statistical methods can help the search for community composition patterns of species and their relationship with available environmental characteristics. For example, predicting the spatial distribution of wildlife populations is an important component of the development of management strategies for their conservation. Understanding abundance of species over time and space helps decision makers geographically target resource allocation at appropriate times in an attempt to remediate problem areas.