股指期货连涨连跌特征的生存分析

发布时间：2018-01-28 01:36

本文关键词： 股指期货生存模型 Gamma分布　出处：《南京财经大学》2014年硕士论文　论文类型：学位论文

【摘要】：伴着我国资本市场的迅速发展,证券资本市场日益扩大的规模,证券市场结构的不断改变,投资者特别是机构投资者对于风险规避以及多样化的投资工具的需求与日俱增,这在客观条件上对股指期货等金融产品的发展与完善助了一臂之力。鉴于股指期货在套期保值、价格发现以及风险管理方面的独特优势,其在国际金融资本市场中己然成为比较成熟的,且是日益发挥重要作用的投资工具。投资者如何有效的投资股指期货?本文从股指期货连涨连跌特征的视角,生存分析的方法来研究分析。本文是从股指期货1分钟高频收益率出发,以技术分析为主,通过生存分析的视角来,从股指期货连涨连跌分钟数以及连涨连跌分钟收益率两个方面来分析股指期货涨跌特征。对于连涨连跌分钟数,我们主要运用了生存模型中的三个主要函数——生存函数、概率密度函数、危险率函数对股指期货指数的连续涨跌分钟数的特征进行较为客观的分析,并计算出期望连涨连跌分钟数为1.8分钟左右,这在理论条件下也是投资者的盈利时间区间。对于连涨连跌分钟收益率,我们进行了Gamma分布的拟合,结果显示股指期货连涨连跌分钟收益率较好的服从Gamma分布,与此同时,我们对连涨连跌分钟收益率的理论均值和标准差和经验的均值标准差进行了对比,发现两者间差异较小,在可接受的范围,这说明对其分布的Gamma拟合还是基本有效的。在Gamma分布的基础上,我们对股指期货连涨连跌收益率区间的概率以及条件概率进行了计算,结果在正文显示；且通过对无条件概率及条件概率的对比发现,其历史收益率在一定程度上对未来收益率会有影响,这说明在我国股指期货市场进行技术分析也是有效的,这也从另一个角度说明了我国金融市场可能还没有达到真正的弱有效市场。最后我们分析了分钟成交量、分钟持仓量变动与连涨连跌收益率的影响,发现这两个因素中只有分钟成交量与连涨连跌分钟收益率呈现出正相关的线性关系,而持仓量变动并没有明显的线性相关关系。通过实证分析得出相关系数之后,文章中也计算了在成交量固定时连涨连跌分钟收益率的无条件概率及条件概率。
[Abstract]:With the rapid development of the capital market in China, the scale of the securities capital market is expanding day by day, and the structure of the securities market is constantly changing. The demand of investors, especially institutional investors, for risk aversion and diversification of investment instruments is increasing day by day. This helps the development and improvement of stock index futures and other financial products under objective conditions. In view of the unique advantages of stock index futures in hedging, price discovery and risk management. It has become a more mature and increasingly important investment tool in the international financial capital market. How can investors effectively invest in stock index futures? This paper studies and analyzes stock index futures from the point of view of the characteristics of continuous rise and fall and the method of survival analysis. This paper starts from the high-frequency yield of one minute of stock index futures, mainly by technical analysis, through the perspective of survival analysis. This paper analyzes the characteristics of stock index futures from the following two aspects: the number of minutes of continuous rise and fall, and the yield of minutes of continuous rise and fall. For the number of minutes of continuous rise and fall, the paper analyzes the characteristics of stock index futures from two aspects. We mainly use the survival function, probability density function and risk rate function of the survival model to analyze the characteristics of the continuous rise and fall minutes of stock index futures index objectively. And calculate that the number of minutes of continuous rise and fall is about 1.8 minutes, which is also the profit time range of investors under the theoretical conditions. For the continuous rising and falling minutes of the yield, we carried out the Gamma distribution fitting. The results show that the stock index futures continue to rise and fall minutes of better yield from the Gamma distribution, at the same time. We compare the theoretical mean and standard deviation with the standard deviation of experience, and find that the difference between them is small and is in the acceptable range. On the basis of Gamma distribution, we calculate the probability and conditional probability of the continuous rise and fall yield range of stock index futures. The results are displayed in the text; Through the comparison of unconditional probability and conditional probability, it is found that the historical rate of return will affect the future rate of return to a certain extent, which shows that the technical analysis is also effective in the stock index futures market of our country. This also shows from another angle that the financial market of our country may not reach the real weak effective market. Finally, we analyze the influence of minute trading volume, minute position change and successive rise and fall yield. It is found that there is only a positive linear relationship between the minute turnover and the continuous rising and falling minutes' yield of these two factors. However, there is no obvious linear correlation between the change of positions. After the correlation coefficient is obtained by empirical analysis, the unconditional probability and conditional probability of the rate of return for continuous rising and falling minutes when the turnover is fixed are also calculated.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F724.5

【相似文献】