Wataru Kobayashi, Director of Chiba University of Commerce Research Center for Economics

Aim of the special feature in "CUC View & Vision No. 51"

Our university is committed to education and research activities as a "comprehensive university of social sciences." Social science can be positioned as an academic field that scientifically analyzes social issues, but what exactly does scientific analysis consist of? Economics textbooks state that the scientific method of analysis is to derive hypotheses through model analysis and then verify them through data analysis. A model is a simplification of complex reality, and is often expressed in diagrams and formulas.

However, such analytical methods are not used in other fields of social science as well. However, that does not mean that model analysis is not used in fields other than economics. Therefore, in this special feature, we ask researchers from multiple fields, including economics, to discuss the position and characteristics of model analysis in their fields. In addition, we will introduce recent trends in simulation analysis using mathematical models and examples of education at our university.

The first article explains the characteristics of model analysis in economics, and then introduces the surplus distribution problem of how to divide the contents of a can as an example of model analysis. Economics has a common knowledge that can be called shared knowledge. The subject matter has expanded rapidly. Models constructed using game theory are constructed for each social phenomenon, and various phenomena can be explained by changing the model. However, in all models, there is a common underlying idea that each economic agent has some kind of value standard and acts rationally based on that value standard, and it is argued that the hidden characteristic of modern economics is its insistence on this idea.

The second article gives an overview of the position and history of model analysis in sociology, and then introduces network analysis as an example of model analysis. Although mathematical model analysis is not the "main" area of current sociological research in Japan, it exists as a trend due to the tradition since the beginning. Early sociology was oriented toward modeling and law-making, and the practice of modeling was an ideology that sociology could be a "science". However, after attempts such as "social systems theory", precise modeling of society was considered to be incompatible with "sociological" research, and it fell out of the mainstream. However, in the process, there were technical problems such as the difficulty of calculation, and this was completely changed by the appearance and development of computers. Network analysis is introduced as a concrete example, and it is also emphasized that it is a model of "relationships" themselves, in the midst of the methodological problems of "methodological individualism (social nominalism)" and "methodological collectivism (social realism)" regarding how to deal with "society".

The third article considers the issue of what kind of model is desirable in order to examine the significance of models in historical theory, and introduces several research studies that are considered to be good examples. According to the author, what is required in historical theory is a model that is free from frameworks that assume rational choice and can explain the complex changes in social systems. As an example of the former, the author introduces a theory on the origins of modern capitalism, and as an example of the latter, a family typology model that applies periphery theory and a model of the ecological history of civilization. Through an examination of these research examples, the author points out that by creating a model, it is possible to make comparisons and obtain general knowledge, and that ``surprises'' in research can be discovered depending on how the model is used.

The fourth article explains the significance and methods of social simulation based on agent modeling, introduces examples of research in archaeology, and also refers to the Club of Rome's "The Limits to Growth." When analyzing and designing social systems, it is difficult to conduct experiments, so computer-based simulation methods are important as a means of overcoming this difficulty. This method can also be used to compensate for the lack of data in fields such as archaeology, which tries to reconstruct human history from material materials. In addition, "The Limits to Growth," published in 1972, used system dynamics simulations, but at the time only a very small number of researchers were able to design and operate the models. In contrast, "2052," published 40 years later in 2012 and predicting the future for another 40 years, has an Excel file available on a website where readers can freely operate the model. There are now many such models that can be used freely by non-experts.

The fifth article introduces the contents of the lecture "Model Simulation" for undergraduate students at our university, and discusses the challenges of course design from the perspective of data science education. In order to perform simulation analysis using mathematical models, mathematical knowledge such as differential and integral calculus and programming experience are essential. However, this lecture does not assume such knowledge or experience, and instead focuses on experiential learning and understanding of the importance of modeling and simulating real-world objects using spreadsheet software such as Excel and various online services. In the first half of the lecture, after introducing many concrete examples, we deal with models such as graphs, linear programming, and queues, and in the second half, we focus on system dynamics and deal with the Malthusian population model, the SIR model of infectious diseases, and the Lorenz equation that expresses chaos phenomena.