【作者】 许自坚；

【导师】 史本山；

【作者基本信息】 西南交通大学 ， 管理科学与工程， 2012， 博士

【摘要】 股指期货是金融市场结构中的重要一环,是金融市场发展和完善的基石。我国于2010年4月16号正式推出了以沪深300指数为标的的沪深300股指期货,在股指期货推出以前,我国股票市场是一个单边交易市场,只能做多不能卖空,市场投资者对市场有较好预期时买入股票,预期较差时卖出股票,容易造成股票市场的大幅波动,使得股票市场的系统性风险很大。股指期货的推出使得卖空机制得以实现,有助于进一步发展和完善我国资本市场,对于建设一个完善的多层次的金融市场体系也有很大的帮助。一般而言,股指期货市场主要有三个方面的功能：套利、价格发现以及套期保值。本文以股指期货为研究对象,实证分析了我国沪深300股指期货的市场特征和市场功能,论文主体包含四个部分,运用的方法模型及得到的主要结论如下所示：论文首先研究了我国沪深300股指期货的价格特征及套利功能。在持有成本模型的基础上运用无套利定价原理,考虑交易费用、冲击成本、卖空限制以及利率变动,推导了我国股指期货的无套利区间。分别以沪深300股指期货的日交易数据和日内5分钟高频数据研究了股指期货的定价误差及影响定价误差幅度的因素。研究表明在我国沪深300股指期货的价格在大多数时间是偏高的,在考虑套利成本的情况下,股指期货的定价在大多数时间是有效率的,但是在股票市场大幅波动的时期,股指期货存在较大幅度的定价误差。从影响股指期货定价误差幅度的因素来看,距到期日越远定价误差越大,现货指数波动越剧烈定价误差越大,股指期货持仓量对定价误差没有显著影响,加息对定价误差的影响跟加息日期有关。当沪深300股指期货存在定价误差时,就可以通过套利获取无风险利润,因此对研究期内的沪深300股指的套利时点和套利利润的大小进行了实证研究,实证结果表明我国沪深300股指期货的定价在研究期内大多是有效的,也就是说在大多数时间不存在套利机会。但是在股票市场大幅波动期间,沪深300股指期货存在正向的套利机会,套利机会主要出现在股指期货推出的前期(IF1005和IF1011),随着股指期货推出时间的延长,套利机会基本消失。第二部分研究了股指期货的价格发现功能。以5分钟高频数据建立向量误差修正模型,显示股指期货与现货指数之间具有长期的协整关系,当因信息冲击发生偏离时,股指期货的调整幅度更为明显,表示股指期货市场在价格发现中要强于现货市场。采用IS和PT模型计算出的价格发现贡献度水平,同样显示股指期货在价格发现中占主导地位。最后辅以方差分解及对脉冲响应反应函数探讨沪深300指数期货与现货指数间信息产生及传递的方式。第三部分研究了沪深300指数与沪深300股指期货的日收益率序列的相关特征。建立Copula-GARCH(1,1)-GED模型实证发现：沪深300指数与股指期货收益率序列之间相关程度非常高,根据不同的标准可选择不同的Copula模型来描述两序列的相关结构；两序列的尾部相关程度非常高,表明当股票市场大幅度波动时,沪深300指数与沪深300股指期货的相关程度显著提高。最后研究了股指期货的套期保值功能。以VaR和CVaR作为套期保值模型的目标函数,推导了基于VaR和CVaR的最优套期保值比例的计算模型,并利用我国沪深300股指期货的实际数据,验证了套期保值的有效性。实证结果发现,基于VaR和CVaR的股指期货套期保值模型能够有效降低投资组合的风险,并能够得到较好的累计收益率。相比较而言,基于CVaR的股指期货套期保值模型能够更好的控制投资组合的风险,基于VaR的股指期货套期保值模型能够得到更优的累计收益率。更多还原

【Abstract】 Stock index futures plays a significant role on financial market structure, it is the cornerstone of financial market development and improvement. In16th April2010, our nation formally launched the csi300stock index futures which was marked to CSI300index. Before stock index futures had been launched, China’s stock market was an unilateral market, which could only buy stocks while could not make a short, the market investors bought the stocks when market had good anticipation and sold them when anticipation worse. These actions might cause a rapid fluctuation, at meantime, cause a enormous systemic risk. Stock index futures realizes a shorting mechanism, helping our country’s capital market structure becomes more perfectly, prompting the formation of an impeccable and multilevel financial market system. Generally speaking, stock index futures market has three main functions: arbitrage, price discovery and hedging, separately. This paper aimed at stock index futures, practically analyzed the characteristics and function of csi300stock index futures market. The dissertation is consisted with four parts; the methods of the model and the main conclusions are shown below:The first part of the dissertation is mainly about the price characteristics and the arbitrage function of csi300stock index futures in China. The no arbitrage pricing theory had been used in the research which was based on The Cost of Carry Model. The transaction cost, the impact cost, the short-selling restrictions and the adjusted interest rates had also been considered; utilizing these factors, the no-arbitrage interval of stock index futures had been deduced. Using stock index futures’daily data and five minutes of high frequency data, the pricing errors and the influence factors of pricing error range had been empirical researched. The studies indicate that the price of csi300stock index futures was on the high side during the most time, if the arbitrage cost had been considered, the price of stock index futures was efficient during the most time, however when a sharp fluctuation happened in the stock market, Stock index futures would have a rapid pricing errors. Looking from the factors, which influence stock index futures pricing error range, the more period from deadline the more error pricing would have, and the more rapid fluctuation of market the more serious pricing error would have. The aggregate gross position of stock index futures had not an apparent effect on the pricing error; the impact to pricing errors, which was caused by raise interest rate, had been decided by the date of the regulation of interest rate.When csi300stock index futures had the pricing errors, investors could get profits without risk through arbitrage. Therefore, this dissertation studied the CSI300stock index arbitrage point and the size of arbitrage profits. The research illustrates that the price of CSI 300stock index futures mostly fluctuate in the no-arbitrage range, stock index futures had not exist arbitrage opportunities and pricing was effective during the most time. But in the stock market fluctuations by a wide margin of period of time, stock index futures cashed and carried arbitrage opportunity, stock index futures existed pricing errors at this time. Arbitrage opportunities mainly appear in csi300stock index futures launched in the early period (IF1005and IF1011), with the stock index futures launched the extension of time the arbitrage opportunities almost disappeared.The second part of the study researched stock index futures price discovery function. The5minutes high frequency data had been used to establish vector error correction model, displayed stock index futures and spotted index had a long Co-integration relationship. When information impact occurred, the adjustment range of stock index futures was more apparent, showing that stock index futures market in price discovery was stronger than the spotted market’s. Then, IS and PT model had been adopted to calculate the information share. The result also showed that stock index futures was in the leading role. Finally, with the variance decomposition and the impulse response function, this part discussed the generation and transmission of information between csi300index futures and spot index.The third part of the dissertation researched the daily return series related characteristics between csi300index and csi300stock index futures. The established Copula-GARCH (1,1)-GED model empirical research shown that the related degree between CSI300index and CSI300stock index futures was very high. According to different standard, there was different Copula model could be chosen to describe two sequences related structures. The two series had high related degree in tail related show that when the stock market wide fluctuating, the correlation between CSI300index and CSI300stock index futures increased sharply.At last, the dissertation researched the stock index futures hedging function. The optimal hedging ratio based on VaR and CVaR method had been deduced, which used Made CVaR and VaR as the hedging objective functions. Then, the effectiveness of hedging with CSI300stock index futures had been examined. The empirical results shown that stock index futures hedging model, which was based on VaR and CVaR, could effectively reduce the portfolio risk, and could also get a better total return. On contrary, stock index futures hedging model, which was based on the CVaR, could better control portfolio risk; stock index futures hedging model, which was based on the VaR, could get better total return.更多还原