This paper introduces how to formulate the CSI300 evolving stock index using the Paasche compiling technique of weighed indexes after giving the GCA model. It studies dynamics characteristics of the Chinese stock market and its relationships with monetary policy intensity, based on the evolving stock index. It concludes by saying that it is possible to construct a dynamics equation of the Chinese stock market using three variables, and that it is useless to regular market-complexity according to changing intensity of external factors from a chaos point of view.
Complexity in the stock market belongs to the global characteristics of the system, and is under the influence of both internal factors, mainly referring to interaction between investment agents, and external factors including the macro-economy, the industrial economy, the regional economy and the company’s development situation. However, it is difficult to use an empirical analysis method to study a certain factor’s influence on the global characteristics of a socio-economic system such as a stock market due to the irreversibility. When a particular factor changes, other factors and system environments change accordingly, so it is hard to discuss the relationships between specific factors and system behaviors while keeping other factors unchanged according to empirical analysis methods.
However, the discrete space model [1] provides an experimental train of thought to study socio-economic systems, which shows the central idea of complexity science, that is, the overall structure can emerge from the restricted local interactions. After Schelling’s research on the segregation of city communities [1], the discrete space model was widely used in many fields, including computer science [2], economics [3], sociology [4], and artificial life [5], and was considered to have significant advantages in the studies of self-organization, phase transitions, and emergent phenomena in complex systems.
Regarding the stock market, the earliest application of the discrete space model was the Artificial Stock Market (ASM), developed by W. Brian Arthur in 1989; its modeling framework has been imitated by some domestic scholars [6], [7] and [8]. However, we did not find that such studies include relationships between stock market behaviors and their influencing factors, and therefore cannot provide effective grounds for decision support.
According to the modeling idea of the discrete network, we constructed a single stock market evolving model based on genetic cellular automaton (GCA) [9], where classifier system put forward by J. Holland [2] is applied to optimize system parameters, then the CSI300 index was formulated so as to study the relationships between the complexity of the whole market and its factors, among which this paper will focus on the relationships between dynamics behaviors of the Chinese stock market and money policy intensity.
Through the relationships between fractal dimension and coefficient of monetary policy intensity analyzed above, we can draw the following conclusions:
(1)
No matter how monetary policy intensity changes, the fractal dimensions of the reconstructed phase space in the Chinese stock market of the A-share is generally between 2 and 3, and their average value is 2.73, which shows that the Chinese stock market of the A-share has fractal characteristics, and three variables are probably enough to construct a dynamics equation for the Chinese A-share stock market.
(2)
To some extent, the administration and regulation of a stock market is equivalent to controlling the complexity of the stock market, rather than directly controlling the stock price or index itself. However, as external factors like monetary policy intensity have nothing to do with the fractal dimension of the Chinese A-share stock market, it may be useless to regulate the complexity from chaos point of view according to changing the intensity of external factors.
(3)
Nevertheless, administration authorities might still count on controlling complexity of stock market according to other external or internal factors such as policy-issuing frequency, message-releasing frequency, investment psychology, and so on. With more relationships between complexity and influence factors being discovered, the means for stock market regulations will become more abundant.
(4)
It is usually the case that government issues monetary policies to intervene in the economy according to commodity prices, employment, growth rate and international balance of payments. When economic problems arise, they usually appear in stock market. So, intervention in the economy looks like to intervene in stock market, and economic goals could be achieved by stock market regulation in the very great deal. However, as its performance in the global financial crisis since 2008, government’s intervention in economy did not bring about the desired efficiency. Facing the problems above, alternative thinking is worth attempting, that is the economy or stock market could be regulated or controlled in the light of complexity characteristics rather than observed data itself.
