Introduction to linear regression analysis
Introduction to linear regression analysis
复制标题
DOI：
10.1080/02664763.2013.816069
复制
发表时间：
2013-10
影响因子：
1.5
通讯作者：
P. B. Ober
中科院分区：
文献类型：
--
作者：
P. B. Ober
作者研究了建模中的一些具体问题，特别是缺失数据、删失数据和截断数据的情况。此外，他们讨论了一系列的分层模型，并考虑到模型的不确定性的参数分布的选择狄利克雷过程的实施。最后，本书以讨论BUGS的语法作为结尾。这本书的主要优点是它将贝叶斯数据分析，MCMC技术和实际实现联系在一起。它是用简单的语言写的，并且包含BUGS代码对于增强读者对书中讨论的方法的理解特别有用。将本书的适用性扩展到函数回归模型将是有趣的，其中协变量可以是函数的。随着越来越多的高维数据，函数协变量允许我们使用微分方程的知识来分析导数。总的来说，我非常喜欢阅读这本书，并认为它给出了贝叶斯建模的简明介绍。它不仅关注许多统计模型的方法学方面，而且还为对BUGS感兴趣的研究人员和研究生提供了完美的实用参考。总之，这本书在很大程度上是一个令人兴奋的和流行的研究领域的大门，它提供了许多实际的见解贝叶斯建模。这是一本文字清晰、结构良好的书，适合各种读者，包括但不限于贝叶斯统计学家、贝叶斯计量经济学家和计算机程序员。因此，它是任何统计参考图书馆的推荐购买。
The authors study some specific issues in modelling; specifically, the cases of missing data, censored data and truncated data. Furthermore, they discuss a range of hierarchical models, and the implementation of Dirichlet processes for taking into account model uncertainty about the choice of a parametric distribution. Finally, the book ends with a discussion of the syntax of BUGS. The main strength of this book is that it links Bayesian data analysis, MCMC techniques and practical implementation together. It is clearly written in plain language and the inclusion of BUGS code is particularly useful to enhance the reader’s understanding of the methodology discussed in the book. It would be interesting to widen the book’s applicability to functional regression models, where the covariates can be functional. With ever increasingly high-dimensional data, functional covariates allow us to analyse derivatives using the knowledge of differential equations. Overall, I enjoyed reading this book very much and thought that it gives a concise introduction to Bayesian modelling. It not only focuses on the methodological aspects of many statistical models, but it also serves perfectly as a practical reference for researchers and graduate students with an interest in BUGS. To summarise, this book is to a great extent a door to an exciting and popular research field, and it provides many practical insights into Bayesian modelling. It is a clearly written and well-structured book for a diverse range of audiences, including but not limited to Bayesian statisticians, Bayesian econometricians and computer programmers. Hence, it is a recommended purchase for any statistical reference library.