Mathematical Sciences: Growth Curve Marginal Models

边际模型是指建立在广义线性模型结构上的模型，但将其扩展到高斯分布以外的广泛的离散或连续值变量分布。本项目将这些模型扩展到特别强调离散数据的增长曲线模型。软件的实施，特别是适度的计算支持将作为这个项目的一部分进行。当在一段时间内观察个体时，数据分析的目的通常是检查个体的变化模式，并检查这些模式在个体之间的共性。这通常会导致个体群体之间的区别，或者将这些模式的特征与群体的特定属性联系起来。本研究将为这类分析制定统计程序。

Marginal Models refer to models built on the structure of generalized linear models, but extended beyond the Gaussian case to a wide class of distributions for discrete or continuous- valued variables. This project will extend these models to growth curve models with particular emphasis on discrete data. Implementation of software, particularly for modest computing support will be undertaken as part of this project. When individuals are observed over a period of time, the objective of the data analysis is usually both to examine the pattern of change for the individuals and to examine the commonalities of these patterns among individuals. Often this leads to distinctions among groups of individuals or to associating features of these patterns with particular attributes of the groups. This research will develop statistical procedures for this kind of analysis.