The Multi-scale DCC-GARCH Model Risk Correlation Study Between Financial Industry

Since the 1980s, with the vigorous development of the financial services industry, the relationship between financial institutions are increasingly close. However, the relationship between financial institutions is complexity, not only promotes the development of the sound of the financial system as a whole, but also brings the possibility of spreading risk in the financial system to infect. In the era of economic globalization of the market economy, the outbreak of the financial crisis is no longer constrained in a certain area, but it have been unable to gauge the speed of the rest of the world. In China, although in domestic internal have not appeared risk caused by the massive financial crisis, but in recent years in the domestic financial system instability phenomenon has aroused people’s vigilance. As domestic financial industry business cooperation between the increasingly frequent, risk correlation between industry is also growing, so in a timely manner to master the financial risk between the industry trend is extremely important.Based on the domestic support of the financial system four big financial industry:the trust industry, banking, insurance and securities industry as the research object, in order to be able to analyze the financial industry is facing the dynamic correlation between risk and industry risk, this paper adopts the wavelet method and multivariate generalized conditional heteroscedastic model with the combination of methods for comprehensive analysis. Through the dynamic relation of multiple time scale analysis, understand the characteristics of the risk associated with between domestic financial industry, clarify the structure of the domestic financial industry association between risk, to be able to more deeply understand the transmission mechanism of systemic financial risk in the early generation, hope at the beginning of the crisis to generate effective early warning to the financial system.Through the empirical analysis finally get the following conclusion:the trust, banking, insurance and securities industries as the backbone of our country finance system four big industry, interrelated relationship between them; And because the risk correlation between industry will changes with the change of the scale, the industry risks due to the change of scale also has different reasons, so the risk control, should be considered in different emphasis on different time scales focusing on control. At the same time through empirical analysis, can be determined in the financial system in China, the bank is the main link, the whole system running from risks associated, once greatly from Banks in turmoil, if in a short period of time can’t solve, may bring great overall crisis of the financial system.This paper is divided into parts:The first chapter is an introduction, mainly elaborated in this section in this paper, choose to four pillars of the domestic financial industry as the research object, investigated the correlation between various industries for the background of the research purpose and meaning. At the same time, also will comb on based on the analysis of research used in the related literature at home and abroad.Fluctuations in the second chapter is mainly about the financial time series model and the introduction of the basic theory of wavelet method and so on, including the ARCH model, GARCH model and the expansion of volatility models in the multivariate DDC-GARCH model. Also include the MODWT transform is introduced, and the poor little Poland and wavelet coefficient of correlation analysis.The third chapter is the empirical analysis part, in this part mainly through to the data samples, the wavelet transform, DCC-GARCH model, granger causality test, finally achieve the financial industry, about four in the dynamic correlation problem under the multi-scale analysis purposes.The fourth chapter is mainly for the empirical analysis this paper summarizes the conclusion.The innovation of this article is on the choice of the research object, select the domestic financial system dominated by four industry as the research object; On the choice of sample data using micro data, choose the representative industry enterprise data weighting of industry problem analysis; Join the wavelet transform in the empirical analysis, under the multi-scale using multivariate generalized conditional heteroscedastic model research dynamic relation problem, to a comprehensive understanding of the data by placing all the information, finally, granger causality test for the correlation between industry clear into to show, so as to ensure the integrity of the empirical analysis.