Applied Smoothing Techniques for Data Analysis
Applied Smoothing Techniques for Data Analysis
复制标题
DOI：
10.1080/00401706.1999.10485676
复制
发表时间：
1999
期刊：
影响因子：
2.5
通讯作者：
R. Deveaux
中科院分区：
文献类型：
--
作者：
R. Deveaux
正如标题所暗示的，这本书提供了数据分析中使用的平滑技术的概述，重点是一维和二维数据。作者的目的是补充现有的书籍，专注于直观的想法和推理的实际问题，而不是估计。这本书共有八章193页，前两章致力于密度估计，后六章（大部分）专注于回归和时间序列的平滑。本文自始至终使用真实数据来说明这些技术。提供了S-PLUS代码来重建大部分的图和分析。对于密度估计和回归，都有一个关于数据探索的介绍性章节，然后是关于推理的章节。第一章介绍了密度估计的思想，重点介绍了飞机设计的六个特征的数据集。然后，使用此数据集和其他数据集作为示例，开发和说明各种技术。这是一种贯穿全书的模式，大部分的推导都保持在最低限度，或者放在边栏中，标记为“数学方面”。这与作者的目标是一致的，即“将非参数平滑介绍给其他科学领域的统计学家和研究人员，他们寻求对该主题的实用介绍”（p. 391）。七)。
As the title suggests, this book provides an overview of smoothing techniques used in data analysis, with emphasis on one-and two-dimensional data. The authors’ aim is to complement the existing books by focusing on intuitive presentation of the ideas and on practical issues of inference rather than estimation.The book consists of eight chapters and 193 pages, with the first two chapters devoted to density estimation and the last six (for the most part) concentrating on smoothing in regression and time series. Real data are used throughout to illustrate the techniques. S-PLUS code is provided to reconstruct the majority of the plots and analyses as well. For both density estimation and regression, there is an introductory chapter on data exploration followed by a chapter on inference. Chapter 1 introduces the idea of density estimation by focusing on a dataset on six characteristics of aircraft design. Various techniques are then developed and illustrated, using this and other datasets as examples. This is a pattern followed throughout the book with derivations for the most part kept to a minimum or put into sidebars denoted “Mathematical Aspects.” This is consistent with the authors’ aim to “introduce nonparametric smoothing to statisticians and researchers in other scientific areas who seek a practical introduction to the topic”(p. vii).