全球股票市场动态特征比较研究——基于STFIGARCH模型

Comparative Study on the Dynamic Characteristics in Global Stock Market——Based on STFIGARCH Model

随着经济全球化的深入发展,资本市场的关联效应不断加强,其中,股票市场的动态特征最为明显。本文使用STFIGARCH模型研究了2003-2012年间中国、美国、英国、日本等股票市场的长期记忆以及非线性、非对称性等动态时变特征。结果表明,就主板市场而言,在全球范围内,英国富时100指数收益率波动通过STFIGARCH模型研究表现出显著的长期记忆性、非线性和非对称性等动态时变特征;而通过研究全球范围内的创业板市场(GEM),发现英国富时AIM创业板指数收益率波动的长期记忆最强,中国创业板指数收益率波动长期记忆特征最弱,中、英两国创业板指数收益率波动的非线性和非对称特征明显。另外,在同一区域内,对主板市场和创业板市场波动特征进行比较,中国相对于其他国家,其创业板和主板的动态特征差异相对较大,而美国对应两个市场的差异则相对较小。这些结论对于政府制定政策、规范证券市场发展以及投资者跨国投资都具有很好的借鉴意义。

With the in-depth development of the global marketization, the correlation of different capital markets is getting stronger,among which the dynamic characteristics of stock market is most obvious one. The stock market plays an important role in the national economy, a close correlation between volatility of stock market and the main body participating in market activities, not only directly affect the fluctuation characteristics of all parties to the in- vestment situation of risk and return, but also for the market regulators issued the policy of macro regulation and control to provide the important reference. In this paper, we apply a STFIGARCH model to analyze the dynamic time-varying character in stock market of China ,America, Britain and Japan from Jan. 2003 to Aug. 2012 ,including long-term memory, nonlinear dynamics, and asymmetry. At the same time, we compare the differences between the Main-Board Market and the Growth Enterprise Market （GEM）. The empirical results are known as follows. First, in the Main-Board Market, asymmetry, long-term memory, and nonlinear dynamics of FTSE100 index are significant, which indicates that the dynamic time-varying characteristics of this board are distinctive. In the stock market there exists a strong long-term memory that means the fluctuation of stock market has persistence. Hence the efficiency of stock market is relatively low, which results in an inaccurate reflection. Both the Main-Board Market and the Growth Enterprise Market of Britain has a strong long-term memory, as we discussed above, this indicates an insensitive fluctuation. This paper studies five large effect countries in global stock markets （or area） were studied,from 2003 January to 2012 August,the UK stock market,America stock market, Japanese stock market,Hongkong stock mar- ket and China mainland stock market as samples, the main board market and the rate of return of the gem index fluctuation characteristics are compared, and then made research with following four aspects ： （ 1 ） in the sample pe- riod, the main stock markets around the world have shown long-term memory dynamic characteristic, nonlinear and non symmetry ,which makes the stock market volatility in the process is persistent, and state transition are ongoing, but conversion speed will vary because of different market. （2） from our global stock market, in the motherboard market, stock return volatility of long-term memory, nonlinear and asymmetry in the UK stock market was the stron- gest. This shows that even in different stock market correlation degree very high volatility, they have much differ- ence, perhaps days and months multiplying various rules and regulations, market and cultural factors in the formation are closely related. （3） from the GEM market in global stock markets of the UK stock market, in the sample period of long memory of the strongest, Shenzhen GEM, the weakest China, in Britain, the GEM market volatility has obvi- ous non-linear and non symmetry. （4） from a single country （ or region） Comparative internal mainboard and carve out board market, in addition to China outside, other countries （or regions） nonlinear motherboard market volatility is significantly higher than that of the country （ or region） studies show that the GEM market,, in the sample coun- tries （or regions）, the difference China the gem and the main board market, in various dynamic characteristics on the maximum, and American corresponding two equity market is the least. This shows that compared USA stock market, the stock market in our country has more emerging stock market characteristics, more need to be perfected in the system. As a result,investors are supposed to highlight these characteristic. Meanwhile, supervisors are ex- pected to guide a long-term effectiveness to promote market development, and observations as well as adjustment are needed, too. Second, in the Growth Enterprise Markets of the world, the highest degree of long-term memory belongs to b-＇YSE AIM UK 50 Index and the lowest to China GEM index, while nonlinear and asymmetric characteristics are significant in the GEM of China and Britain. When impacted with information, the transformation speed of these two states depends on the degree of nonlinear, the higher degree is, the faster the speed is, and also the higher the uncer- tainty is. The degree of nonlinearity of the Growth Enterprise Market （GEM） in both China and Britain is relatively high, which indicates a complicated and variable market, also a high risky market. Therefore investors are expected to be more cautious. On the contrary, the degree of nonlinearity of the Main-Board Market and the Growth Enter- prise Market is lower so when impacted with information, the stock market will smoothly transform one state into an- other, in other word, it has a better stability. Third, in the paper, we also study the differences between the Main- Board Market and GEM in same regions. The results show that there are many differences in China Stock Market, while the differences in America are very few. This is probably because China Stock Market lies in a growth period now, so the regime is incomplete yet. vestors＇ behavior and the supervision Especially the Growth Enterprise Market just emerged in recent years, both in- efficiency of supervision authorities need to be perfected. So market partici- pants of all kinds are expected to positively formulate laws and vide a finance platform to national economic. These conclusions regulations, and perfect China Stock Market to pro- have much reference for the government policies to standardize securities market development and are helpful to the transnational investment.