| Financial markets are complex systems with many-body interactions. Since these sys-tems are similar to the traditional physical ones, more and more physicists put effort in-to investigating their properties. The financial systems undergo rapid development, and a large amount of historical market data have piled up. These data offer the possibility to the study of the systems with physics, particularly with statistical physics, and contribute to the birth of econophysics. Different from the traditional finance, econophysics pays more attention to temporal and spatial correlations, many-body interactions, microscopic gener-ation mechanisms etc. During the 20 years of the development of econophysics, physicists have made a lot of analyses and obtained plenty of results.The most basic property of a financial system is the fat-tail distribution of the price returns. From the view of physicists, the dynamic behavior and community structure of the system can be respectively characterized by temporal and spatial correlation functions. The price volatilities are long-range correlated in time, which is called volatility clustering. In this work, we focus on the microscopic originations of the return-volatility correlation and sector structure, which are respectively well-known temporal and spatial correlations, and also the generation mechanisms of the volatility-return correlation nonlocal in time. Chapters 2,3 and 4 are the original content of this work. The innovations of this work:1. Based on agent-based modeling and the investors’ behaviors in real markets, we respec-tively introduce asymmetic and multi-level interactions to investigate the return-volatility correlation and sector structure, and reveal the microscopic originations of them.2. We pro-pose effective methods to determine the key parameters from historical market data, rather than from artificially setting. It improves the reliability of our models, and gives physi-cal meanings to the parameters, which enables them to characterise properties without the models. Other parameters are also calculated or estimated from the empirical findings.3. We explore the generation mechanisms of the volatility-return correlation nonlocal in time at the macroscopic and microscopic levels.In Chapter 1, we introduce the origin and development of econophysics simply, and lay emphasis on the obtained results, including the temporal and spatial correlations, as well as some microscopic models. Next, we display the motivation and content of this work.In Chapter 2, we focus on the return-volatility correlation in stock markets, i.e. the correlation between the past returns and future volatilities. The correlation is the leverage effect if it is negative, and is the anti-leverage effect if positive. These two effects are particularly important for understanding the price dynamics. However, the microscopic origination of them is still not understood. To study the return-volatility correlation, we take into account the individual and collective behaviors of investors in real markets, and construct an agent-based model. The agents are linked with each other and trade in groups, and particularly, two novel microscopic mechanisms, i.e., investors’asymmetric trading and herding in bull and bear markets, are introduced. Further, we propose effective methods to determine the key parameters in our model from historical market data, rather than from statistical fitting of the results. For six representative stock-market indices in the world respectively, the model produces the corresponding leverage or anti-leverage effect, and the effect is in agreement with the empirical one on amplitude and duration. At the same time, our model simulates other features of the real markets, such as volatility clustering and the fat-tail distribution of returns. After further analysis, we reveal that for the leverage and anti-leverage effects, both the investors’asymmetric trading and herding are essential generation mechanisms.In Chapter 3, we consider the sector structure in stock markets, which is an important spatial correlation. However, the microscopic origination of the sector structure is not yet understood. We introduce a multi-level herding mechanism in constructing an agent-based model to investigate the sector structure combined with volatility clustering. According to the previous market performance, agents trade in groups, and their herding behavior comprises the herding at stock, sector and market levels. Further, we propose methods to determine the key model parameters from historical market data. From the simulation, we obtain the sector structure and volatility clustering, as well as the eigenvalue distribution of the cross-correlation matrix, for the New York and Hong Kong stock exchanges. These properties are in agreement with the empirical ones. Our results quantitatively reveal that the multi-level herding is the microscopic generation mechanism of the sector structure, and provide new insight into the spatio-temporal interactions in financial systems at the microscopic level.In Chapter 4, we focus on the volatility-return correlation nonlocal in time. It is impor-tant, since the local volatility-return correlation is zero. We construct a macroscopic model and a microscopic model to investigate the mechanism of volatility-return correlation non-local in time. First, we build a macroscopic model with coupled interactions of short-term volatility and long-term volatility. With model parameters chosen appropriately, our model regenerates the non-zero volatility-return correlation nonlocal in time. To further under-stand the microscopic origin of the correlation, we introduce an agent-based model with a novel mechanism, that is, the asymmetric trading preference in volatile and stable mar-kets. The simulation results are in agreement with the empirical ones, indicating that the correlation arises from the asymmetric trading preference in volatile and stable markets.In Chapter 5, we summarize the works in Chapters 2,3 and 4, and bring on the future prospect.
|
PDF Full Text Request