An S-Plus Functional Data Analysis Module

功能数据出现在许多医学研究领域，例如对生长模式、步态、黑色素瘤发病率、CD4细胞计数和许多其他领域的研究。事实上，任何一组随时间（或空间）收集的测量，包括生存分析中的时间相关协变量，都可以被认为是功能数据。将这些数据视为函数而不是不相关的点有许多优点，也许最重要的是能够将派生信息常规地包含到分析中。从历史上看，功能数据一直使用多变量或时间序列方法进行分析，但这些方法不适用于不规则间隔的数据或不同受试者在不同时间测量的数据。最近的进展使分析功能数据成为可能，在这里，我们建议实现一个S-Plus模块，用于功能数据分析。该模块将是Ramsay和Silverman（1997）开发的探索性方法的商业实现，具有许多扩展，包括广义线性模型，生存分析和非线性最小二乘模型的新方法，以及扩展到具有不同基的函数。新模块将无缝集成功能数据分析方法到S-Plus。建议的商业应用：随着计算机融入日常生活，研究人员收集功能数据的能力变得越来越普遍。目前还没有可用于处理功能数据的商业产品。与现有技术相比，所提出的方法具有显著的优势。一个精心设计和全面的方法来实现这些模型将找到一个现成的市场。

Functional data arise in many fields of medical research, with examples from studies of growth patterns, gait, melanoma incidence rates, CD4 counts and many other areas. Indeed, any set of measurements gathered over time (or space), including time dependent covariates in survival analysis, may be thought of as functional data. There are many advantages of viewing such data as functions rather than disconnected points, perhaps the most important being the ability to routinely including derivative information into the analysis. Historically, functional data has been analyzed using multi-variate or time series methods, but these methods do not work well for irregularly spaced data or data measured at different times for different subjects. Recent advances make it possible to analyze such data as functions Here we propose to implement an S-Plus module for functional data analysis. This module will e a commercial implementation of the exploratory methods developed by Ramsay and Silverman (1997), with many extensions, including new methods for generalized linear models, survival analysis, and non-linear least square models, and extension to functions with different bases. The new module will seamlessly integrate functional data analysis methods into S-Plus. PROPOSED COMMERCIAL APPLICATIONS: As computers become integrated into daily lie, the ability of researchers to collect functional data is becoming more common. There are currently no commercial products available for handling functional data. The proposed methods have significant advantages over existing techniques. A well designed and comprehensive method for implementing these models will find a ready market.