Reviewed by
Bogumil Kaminski
Warsaw School of Economics
The book Economic Foundations for Social Complexity Science: Theory, Sentiments and Empirical Laws, published by Springer in 2017, was edited by Yuji Aruka and Alan Kirman, who are both seasoned scholars in the field of social complexity modelling. It is a follow-up publication to International Conference on Socio-economic Systems with ICT and Networks that took place in March 2016.The editors undertook an effort to create a collection of thirteen papers that present a range of perspectives highlighting that considering the economy as a complex adaptive system may lead to different conclusions than standard modern macroeconomic analysis. However, be warned that the book does not present any ABM models. A typical JASSS reader should consider it as a potential source of ideas, concepts and empirical evidence that can be used to guide their development. The book is organised around three major themes: (1) theoretical foundations of looking at the economy as a complex system, (2) empirical investigations of complex networks and sentiment analysis, and (3) empirical analysis of financial markets. All these aspects are very well summarised in an Introduction chapter by A. Kirman. I found this chapter interesting to read on its own, in particular giving a good historical perspective on the subject.
The first part of the book, “Theoretical Foundations”, is opened by the paper by Y. Aruka on systemic risks analysis using an input-output perspective to the economic system. A distinguishing point of view of this work is an extensive discussion of immunology in biological systems and how its understanding can guide the study of stability of economic systems. The next chapter, by A. Chatterjee, A. Ghosh and B.K. Chakrabarti, focuses on the analysis of Gini and Kolkata indices as measures of socioeconomic inequality, illustrating their relationship using diversified empirical data. J.B. Roser and M.V. Roser in Chapter 4 provide a theoretical discussion on the evolution of economic institutions. This paper is a good reference for a reader wanting to get an understanding which are the main streams of thought of behavioural and institutional economics related to this subject. Using a similar approach, S.-H. Chen and R. Venkatachalam, in Chapter 5, provide a discussion of a historical overview of agent-based modelling approaches from a perspective of how networks were represented in them. Next, J. Mimkes discusses the applications of Stokes integrals to various areas of economics, finishing with the discussion of empirical data on the growth of the Japanese economy. This part is concluded by the work of B. Dürning, N. Georgiou and E. Scalas focusing on the analysis of wealth distribution. They propose an interesting model that is analysed using standard tools from mathematical economics domain. However, I think that it can give an inspiration for the development of more complex agent-based models.
The second part of the book, titled “Complex Network and Sentiments”, switches to empirical papers. In Chapter 8, M. Kubo, H. Sato, A. Yamaguchi and Y. Aruka, using Latent Dirichlet Analysis technique, uncover employment trends in Japan. The proposed technique of an analysis of textual data can prove a useful tool also in other applications, and I think it can be successfully used by ABM researchers when calibrating their models against unstructured data. E. Liu, T. Ito, K. Izumi, K. Tsubouchi and T. Yamashita in Chapter 9 present an application of a standard word2vec model for textual data analysis in the scenario of source documents prepared in two languages. Their analysis of English and Japanese texts shows that in order to be able to obtain good mappings of models between two languages, the original multilingual text data should be similar. The last chapter of this part, presented by K. Izumi, H. Suzuki and F. Toriumi, proposes a method for empirical analysis of the impact of a significant external shock (Great Eastern Japan Earthquake in their case) on information flows between stocks and index futures. Their findings can give a valuable inspiration for the development of ABM models of financial markets that would be able to reproduce the observed market reactions to external shocks.
The last part, dedicated to “Empirical Laws in Financial Market”, is opened by the work of K. Sharma, S. Shah, A.S. Chakrabarti and A. Chakraborti, reviewing techniques that allow extracting information from stock market data. The paper applies them to daily information from Bombay stock exchange and is a good reference of standard methods stemming from econophysics that can be used for such analyses. In Chapter 12, M. Miyano and T. Kaizoji empirically analyse the divergence between a company’s fundamental value and its share price. This kind of divergence is often studied by ABM researchers, so their results can be an interesting reference for empirical studies. The book concludes with a paper that investigates the structure of banks’ balance sheets, published by K. Fukuda and A.-H. Sato. They identify stylised facts that can be used by ABM researchers, especially working in the field of banking sector stability and its susceptibility to cascading failure.
In summary, the reviewed volume consists of thirteen papers investigating various avenues of non-mainstream thinking in economics. I think that it can be interesting to two kinds of readers. First – when taken as a whole – it offers insights into trends of such analysis, which would be valuable for scholars wanting to get an understanding of which topics currently attract research interest in this area. Second, readers focusing on specific areas of ABM modelling (e.g., wealth distribution, financial markets or banking sector) can find in the selected chapters theoretical concepts and empirical evidence that can help them better design the created models.
Return to Contents
of this issue