Probability Models in Engineering and Science
Probability Models in Engineering and Science
复制标题
工程和科学中的概率模型
DOI：
10.1198/tech.2006.s441
复制
发表时间：
2006
期刊：
影响因子：
2.5
通讯作者：
A. Esmaili
中科院分区：
文献类型：
--
作者：
A. Esmaili
随着主题的进展，先决条件在水平和深度上有所不同，从概率（有时是测度论）和马尔可夫链到统计推断（在Bickel和Doksum 1977的水平）和模拟方法。然而，根据个人的背景和兴趣，对HMM感兴趣的理论家和实践者可以获得一些有用的材料。在章节的结尾没有练习，但是专著在许多章节的结尾用一个很好的文献综述和补充材料来丰富。因此，它适合独立学习，许多章节可以很容易地纳入随机过程或计算统计的高级课程。即使有多个作者（除了列出的作者外，贡献者还包括R。杜克角P. Robert，G. Fort和S. Boucheron），风格统一，值得称赞。正文中几乎没有印刷错误。有缺陷的索引（页码不匹配）已由作者纠正（见http：//www.tsi.enst.fr/cancappe/ihmm/errata.html）。这本专著是一种宝贵的资源。它提供了一个很好的文献综述，一个优秀的国家的最先进的研究所需的理论和算法，并充分说明了众多的应用HMM。它远远超出了HMM的早期资源-MacDonald和Zucchini（1997）对离散值HMM的基本介绍以及Elliott，Aggoun和摩尔（1995）强调工程的概率帐户。理论家将受益于理论基础的严格发展，而那些以应用为导向的兴趣将喜欢算法和应用的清晰介绍。隐马尔可夫模型方法将继续存在，并将比以往任何时候都更快地发展，本专著将作为这方面的标准参考。封底指出：“它将对统计学、信号处理、通信工程、控制理论、计量经济学、金融等领域的研究人员和从业人员有用。. . .”这种说法是完全合理的，我预计这项工作将在未来几年内为许多技术计量学读者服务。
The prerequisites vary in level and depth as topics progress, ranging from probability (sometimes measure-theoretic) and Markov chains to statistical inference (at the level of Bickel and Doksum 1977) and simulation methods. Yet, depending on one’s background and interests, there is some useful material that is accessible to theoreticians and practitioners interested in HMM. There are no exercises at the end of chapters, but the monograph is enriched with a good literature review and complementary material at the end of many chapters. Thus, it is suitable for independent study, and many of the chapters can be easily incorporated into advanced courses on stochastic processes or computational statistics. Even with multiple authors (in addition to those listed, contributors include R. Douc, C. P. Robert, G. Fort, and S. Boucheron), the style is uniform and commendable. Very few typographical errors were noticed in the main text. The flawed index (with mismatched page numbers) has been rectified by the authors (see http://www.tsi.enst.fr/~cappe/ihmm/errata.html). This monograph is a valuable resource. It provides a good literature review, an excellent account of the state of the art research on the necessary theory and algorithms, and ample illustrations of numerous applications of HMM. It goes much beyond the earlier resources on HMMs—the basic introduction to the discrete valued HMM by MacDonald and Zucchini (1997) and the probabilistic account with engineering emphasis by Elliott, Aggoun, and Moore (1995). Theoreticians will benefit from the rigorous development of the theoretical underpinnings, and those with application-oriented interest will like the clear presentations of the algorithms and applications. The HMM methodology is here to stay and will grow more rapidly than ever, and this monograph will serve as a standard reference in that context. The back cover states that “it will be useful for researchers and practitioners in areas such as statistics, signal processing, communications engineering, control theory, econometrics, finance. . . .” This claim is fully justified, and I anticipate this work to serve well many Technometrics readers in the coming years.