Research On The Influence Of Investor Sentiment On The Basis Of China's Stock Index Futures

The economic crisis of the 1970 s led to the drastic fluctuations of stock markets in western countries.Stock index futures with the mission of hedging were first introduced in the United States in 1982.Since then,stock index futures experienced five years of rapid development.After U.S.stocks market fell nearly 25% on Oct.19,1987 which caused global stock markets reeling,stock futures were blamed for contributing to the crash,regulators tightened oversight of stock futures and set new guidelines.Stock index futures suffered heavy losses during this period.After a short period of stagnation,stock index futures regained vitality in the 1990 s,and stock index futures markets were set up in many countries.Not until April 16,2010,China's first stock index futures CSI 300 stock index futures showed up.In view of the outstanding performance of CSI 300 stock index futures after listing,CFFEX launched SSE 50 stock index futures and CSI SmallCap 500 stock index futures on April 16,2015.But in the same year,during the crash the stock index futures falling more than the stock index,a sharp discount phenomenon appeared,public opinion was that the stock index futures market manipulated by malicious short sellers bringing down the stock market in turn.Up to the end of 2018,the stock index futures was still at a limited transaction status.In the following three years after the stock market crash,the stock index futures market had been operating in a state of negative basis for a long time,and there is no consensus about what causes this phenomenon.The existing literatures on the influence factors of basis can be roughly divided into four categories: dividend?interest rate and tax,liquidity and arbitrage,market volatility and left-tail risk,and investor sentiment.Most of the early literatures were based on the theory of perfect market and no arbitrage,believing that futures prices should meet the Cost of carrying Model(COCM),and the basis could be explained by dividend,interest rate and tax,such as Cornell and French(1983).However,the performance of the basis was inconsistent with the theoretical value given by COCM.Scholars began to expand the assumptions of perfect market and no arbitrage,and then studied how to explain the behavior of the basis in imperfect market.At this stage,scholars found that both arbitrage and liquidity were closely related to basis,such as Kumar and Seppi(1994),Roll and Schwartz(2005).Other scholars,such as CHEN,et al(1995),Han and Liang(2019),believe that the spot market volatility and left-tail risk have a significant impact on the basis.In recent years,from the perspective of investor sentiment in behavioral finance theory,researches on the basis behavior have also emerged quietly,such as Zheng and Ling(2015).There are relatively few literatures about investor sentiment's influence on the stock index futures basis,which can be roughly divided into three categories.The first category use a single index to represent emotion,such as the risk premium of skewness measuring investors' left tail risk aversion.The second category adopt comprehensive investor sentiment indicators for research.Most of these articles are based on the ideas of Baker and Wurgler(2006).The third category are to study investor sentiment through social media.However,scholars have rarely studied how the mutual contagion of futures-spot markets sentiment affects the stock index futures basis.Therefore,empirical research is conducted on the basis and mispricing of stock index futures by using the data of Chinese market since February 17,2016 to November 30,2018 through four steps.The first step is to construct investor sentiment in futures market through principal component analysis.Second,the SVAR model is used to identify the shock of spot market volatility,investors' lefttail risk aversion and investor sentiment on investor sentiment.The third step is to obtain the cumulative impact time series of each shock on investor sentiment during the empirical period through the Historical Decomposition method.The fourth step is to study the nonlinear influence of investor sentiment changes caused by various shocks on the stock index futures basis and mispricing through Markov switching model.This paper draws four main conclusions :(1)the volatility of the spot market has a positive impact on investor sentiment,and the investor sentiment change caused by its shock has a positive impact on the basis and mispricing of stock index futures.(2)investors' left-tail risk aversion negatively affects investor sentiment,and the investor sentiment change caused by its shock positively affects the basis and mispricing of stock index futures.After 2015's crash,investors in the futures market gradually eased their concerns about the left tail risk,but when the market started to drop continuously,investors in the futures market gradually increased their concerns about the left tail risk.(3)investor sentiment is continuous,and the change of investor sentiment caused by its own shock has a positive impact on the basis and mispricing of stock index futures.(4)during the period of futures market irrationality,the cross-market emotional transmission has a greater impact.This paper has three main innovations :(1)the original investor sentiment represent index used is different from the previous literatures.(2)by means of the Structural Vector Autoregression model(SVAR),the mutual contagion of futuresspot markets sentiment is studied,and the shocks of the spot market volatility,investors' left-tail risk aversion and investor sentiment on investor sentiment are identified.(3)the Historical Decomposition method and Markov switching model are also used to study the nonlinear effects of the shock of the spot market volatility,the investors' left-tail risk aversion and the investor sentiment on the basis and mispricing of stock index futures.The shortcomings are as follows:(1)due to the complexity of empirical research and the long time interval,we can only get the results in the mean sense of the whole interval,but fail to make more analysis on the subintervals,which may neglect many details with rich meanings.(2)The constraction of investor sentiment is not accurate enough to fully reflect investor sentiment.In the future,we could improve the construction of investor sentiment indicator,and research more detailly.