Abstract

The first stock index futures contract began trading in 16 April 2010 in China, and this contract based on HuShen 300 stock index. The study about index futures and spot index are becoming focus. In China, financial microstructure theory research about index futures is just beginning. This study divided the relationship between index futures and spot index markets into two aspects according to the microstructure theory: the relation between stock market volatility and the introduction of the stock index futures, and the relation about comparison of the pricing efficiency between the index futures markets and the spot markets.

A few previous studies did pay adequate attention to the distinction due to different empirical method. The using and comparison of diversified methods lead to gain more explicit and credible result. And the credible empirical results are very close to the size and selection of the sample, the sample employed should be comparable and ample size. There are not empirical studies using the data of HuShen300 index futures in abroad researches and the data of simulation or short-term was employed for the studies in domestic. And a few studies have used the diversified methods and samples simultaneously. The study analyzed deeply the microstructure relation between the two markets through the two aspects above employed diversified empirical methods and comparative method. The daily data and the intraday data（1-min, 5-min）of the Hushen 300, Hang Seng, H-shares, Nikkei 225（SGX）, and the A50 index&index futures over the period 2010 to 2011 are used as the empirical samples. Empirical evidence confirms the theoretical hypothesis in chapter one and gives prominence to the features of HuShen 300 index futures markets by comparison.

The first relation is composed of the difference between spot price volatility before and after the introduction of index futures and the relation between the two markets at index futures expiration. The study examines the relation between the stock market volatility and the introduction of index futures employed parametric, non-parametric test and modified ARCH models and examines the relation between the stock market volatility and the major events in the same period. On the basis of comparison of the two tests, the study gives more credible evidence of the causality on the spot volatility and the futures. There are not unanimous conclusions about the relation. The empirical evidence did not support that there are any causalities between the stock market volatility and the introduction of index futures, as well as the introduction of index futures is a sufficient and necessary condition of the stock market volatility. The empirical analysis employed Hushen 300 sample shows that the introduction of index futures increases the spot markets volatility in the short term, increases volatility but not significantly, and in the long term decreases volatility expelling other factors which can interfere with the observation.

The study provides the evidence on the expiration effects in the Hushen 300, Hang Seng, H-shares, Nikkei 225（SGX）, and the A50 index markets. Empirical analysis this part applies the parametric, non-parametric test of the spot trading volume rate, relative volume, price reversal index, and volatility series at index futures expiration and nonexpiration and spot trading volume, volatility series models. The part devises innovatively the index of price reversal level. The results reveal that the expiration effects ought to include the spot trading volume effects and the spot price volatility effects. Some spot markets exist abnormal changed in trading volume or in price volatility, others in both. Empirical results reveal not only a significant increase in spot trading volume, but also the existence of a significant increase in spot volatility in H-shares spot index at index futures expiration. There are not significant expiration effects in Hushen 300, Hang Seng, H-shares, Nikkei 225（SGX）, and the A50 spot markets. The dissimilarities in futures contract settlement institution between these markets lead to the difference.

The second relation is composed of the information transmission between the two markets, the comparison of the information contents and operational efficiency（liquidity）between the two markets. The study provides an innovative analysis framework on intraday dynamic relation between index futures and spot markets which must consist of the interaction on nonsynchronous and synchronous trading hours. In each stage, the relationship of the first and second moment data should be included. This part adopts minutely high-frequency data of returns and volatility for 5 index futures and spot index markets, because that daily data does not have the ability to statistically assess such relation in most studies. The Granger test, contemporaneous lead-lag relation models, correlation test, equality test of overnight returns, VAR model and VEC model are applied to analyze the lead-lag relationship between the two markets in the intraday nonsynchronous trading stage and synchronous stage. Empirical results display that the futures price volatility before the spot market opening can predict statistically the spot price volatility after the spot market opening over 15 and 30 minnutes in all market samples. Meanwhile, the spot price volatility before the spot market closing can predict statistically the futures price volatility after the spot market closing in all markets sample. The empirical results also show that the difference estimate between the using of the volatility data and the returns in the nonsynchronous trading stage. Empirical results in the synchronous stage confirm generally that futures market plays a price discovery role, implying that futures prices contain useful information about spot prices even when the spot price leads the futures price. The lead-lag relationship in the Hushen 300, Hang Seng, H-shares markets is significant but indeterminate in the Nikkei 225（SGX）and the A50 markets. The results of the synchronous stage imply the difference between the native markets and the non-native markets as well.

The Garbade-Silber model, the impulse response function test and the variance decomposition method are employed to analyze the comparison of the information contents in the returns（volatility）between the futures and spot markets. Empirical results show that futures prices contain more useful information about spot prices than the spot prices themselves. The information contents of Hushen 300 index futures prices contain slightly less information than the Hang Seng and H-shares index futures but significantly more than the A50 index futures. The Hushen 300 spot index prices have strong influence on the H-shares index futures and particularly on the A50 index futures prices in the opposite direction.

The ARMA model and the multivariate regression model contained the futures variable are applied to forecast the spot index price. Forecast results confirm the conclusion above.

Finally, the study examines the relation of operational efficiency（liquidity）comparison between the index futures and the spot markets using the trading volume indicators, variance ratio test, execution costs model and market depth method. Different methods have different results. The results using the trading volume indicators show the liquidity of the futures markets is greater than the spot markets except the A50 futures. The results using variance ratio test and execution costs model display that the liquidity of the futures markets is greater than the spot markets only in the Hushen 300 markets and an inverse result in others markets. These differences result from the method selection, the level of volatility, short sales constraints, the variety of derivatives and the site of the markets. These spot markets chosen as samples have higher liquidity too.

Keywords: Index Futures Spot Index Volatility Expiration Effect Pricing Efficiency