A Research On The Dynamic Thresholds Of CSI 300 Index Future Deviation Based On Thresholds Autoregression Model

Since the standardized future transaction was introduced from market, the function of price discovery appeared between the spot market and the future market. Researchers and dealers gradually accepted this characteristic and made it a significant function of the future market by applying to research and transaction. According to the theory of stock index future, the long-term equilibrium between the stock index and its future comes from the mechanism of arbitrage. Basis, which fluctuates following the stock index, is the base of risk aversion and arbitrage transaction, and is really important for related research. However, the mechanism of going long or short in China’s stock index future market is asymmetric and immature, which makes the mean reversion of basis time-varying and non-linear.The main research object of this paper is the series of non-dimensional basis, which aims at eliminating the price distortion caused by the widely fluctuating of stock index to give an objective analysis. The data used in this paper starts at April 16th.2010 when CSI 300 stock index future formally launched, and ends at August 28th,2015 when the stock index future market was substantively’closed. The daily data of stock index and future index is selected to calculate the non-dimensional basis by applying the cost of carry model. Then under the frame of threshold autoregression model a quantitative analysis is raised by using the threshold property to characterize the heterogeneous fluctuations inside and outside the arbitrage-free interval. The empirical analysis concludes the property of non-dimensional basis and the heterogeneity of China’s stock market at different stages, which helps to distinguish between China’s and mature stock market and find out the early signs of abnormal stock market fluctuations like 2015.Based on basic statistical analysis and contrast analysis of static and dynamic thresholds, the following conclusions have been obtained. First, the arbitrage-free interval can be revealed only by the series of basis, which means that it is an objective existence generated spontaneously from transactions. Second, compared with foreign mature stock markets, the short-term characteristics in China’s market are more obvious. Besides, the price discovery function between the futures and the spot market has a certain self-repair effect after being deviated from the relatively large shock, but the self-repair is limited and needs some time. If the market is shocked significantly again during the repair period, it is likely to lead to market failure. According to the empirical conclusions, this paper gives the corresponding policy recommendations, which are expected to provide a further understanding of China’s stock index future market for the regulatory, financial institutions and investors.