Self-organised criticality: theory, models and characterisation
Self-organised criticality: theory, models and characterisation
复制标题
自组织临界性：理论、模型和表征
DOI：
10.1080/02664763.2014.913844
复制
发表时间：
2014
影响因子：
1.5
通讯作者：
Y. Laberge
中科院分区：
文献类型：
--
作者：
Y. Laberge
介绍时间序列和/或分析财务数据。它也将是有用的人与金融数据的统计知识很少。这本书以一个非常简短的介绍金融数据的一些概念-回报率，收益率，波动性-和一个非常简短的介绍R。我发现金融的介绍很有用，而R的介绍可能太短了，不熟悉R的读者可能想用许多像样的介绍性文本之一来补充这一点。R在整本书中都被用来完成示例，R代码通常包含在本节的最后。这些代码段对于只有R基础知识的读者来说非常有用。第1章还包含了一些基本概率的主题（分布函数，矩，一些特定的分布）。这应该是任何统计学家都熟悉的材料。第2章对线性时间序列（自回归移动平均模型，并简要提及长记忆过程）进行了非技术性介绍。第三章将介绍三个案例研究。这两章中的材料或多或少与人们期望在任何“时间序列介绍”一书中找到的相同。这些例子对于以前没有见过时间序列的读者来说非常有用。在第4章中，Tsay介绍了波动率模型，如自回归条件异方差（autoregressive conditional heteroscedastic，简称ARCH）和广义自回归条件异方差（generalized autoregressive conditional heteroscedastic，简称GARCH）过程，以及这些思想的一些变体。虽然现代时间序列教科书中经常涉及到时间序列和Gynecological过程，但这里的信息比我们通常看到的要多。同样，这是一个非技术性的介绍，在第5章中有更详细的例子。最后两章对统计学家来说就不那么标准了。首先，有一章是关于高频数据的分析，实际交易发生在不规则的时间间隔，价格变化是离散的，通常为零。重点是价格变化模型和下一次交易时间的自回归条件久期模型。接下来是一章关于风险度量的内容，如风险价值和预期缺口。总而言之，Tsay的介绍对于任何需要快速介绍财务数据统计分析的人来说都是一本有用的书。它很容易阅读，一个对基本统计和R有一点了解的读者在理解材料时不会有什么困难。
introduction to time series and/or the analysis of financial data. It will also be useful to people with little knowledge of statistics working with financial data. The book starts with a very short introduction to some concepts of financial data – returns, yields, volatility – and a very short introduction to R. I found the introduction to finance useful whereas the introduction to R is probably too short, and a reader unfamiliar with R may want to supplement this with one of the many decent introductory texts available. R is used throughout the book to work through examples, with R code included typically at the end of the section. These segments of code will be very useful for a reader with just a basic knowledge of R. Chapter 1 also contains a few topics from basic probability (distribution functions, moments, some specific distributions). This should be familiar material to any statistician. Chapter 2 gives a non-technical introduction to linear time series (autoregressive moving average models with a brief mentioning of long-memory processes). This introduction is followed up in Chapter 3 with three case studies. The material in these two chapters is more or less the same as one would expect to find in any “introduction to time series” book. The examples will be very useful for a reader who has not met time series before. In Chapter 4, Tsay introduces volatility models, such as autoregressive conditional heteroscedastic (ARCH) and generalized autoregressive conditional heteroscedastic (GARCH) processes and a number of variations of these ideas. Though ARCH and GARCH processes are often covered in modern textbooks on time series, there is more information here than we usually see. Again, it is a non-technical introduction, which is followed up in Chapter 5 with more detailed examples. The last two chapters are much less standard material for a statistician. First there is a chapter on the analysis of high frequency data, where actual trading takes place at irregularly spaced times and price changes are discrete and often zero. The focus is on models for price changes and autoregressive conditional duration models for the time to the next trade. This is followed by a chapter on risk measures such as value at risk and expected shortfall. All in all, Tsay’s introduction is a useful book for anyone needing a quick introduction to statistical analysis of financial data. It is easy to read, and a reader with a little knowledge of basic statistics and R will have little difficulty understanding the material.