中国股市中小板市场反转效应实证研究

发布时间：2018-04-21 08:34

本文选题：反转效应 + 中国中小板　；参考：《西南财经大学》2013年硕士论文

【摘要】：自从六十年代Fama提出了有效市场理论以来,学术界对有效市场的存在性一直争论不休。有效市场理论认为：在该市场中,存在大量理性并且想要利益最大化的投资者,他们享有共同的信息,并且他们解读信息的方法是相同的,当前的股票价格已经完全反映了市场中所有的可得信息。这一理论在20世纪70年代尤为盛行,一度被学界尊为圣经。但是进入80年代以后,不断有学者发现某些股价运动现象难以用有效市场理论解释。1985年Debont和Thaler (1985)提出反转效应,是指在某一时间点,根据过去一段时间一系列股票的收益率,对它们进行排序,把表现最差的几十家股票称为输家组合,而表现最好的几十家称为赢家组合。随后从中长期(2到5年)考察他们之后的表现,他们发现发生了反转,输家组合的收益率表现好于赢家组合。而这一现象被称为反转效应。最早由Jegadeesh和titamn (1993)提出的动量效应,又称为惯性效应,是指：按照反转效应中描述的方法排序。在接下来的短期内(3到12个月),赢家组合的收益率高于输家组合。而通过买入赢家组合,卖出输家组合,构建零投资组合,就能获取超额回报,而这种投资策略被称为动量策略。笔者出于探究本国反转效应和动量效应规律的目的,选取了中国中小板市场作为自己的研究对象。由于中小板市场成立已经近十年时间,股票数量增加至七百余家,对于研究短期反转效应来说已经积累了足够的样本数量。而本文的研究意义主要有以下几点：第一、从学术角度来看,目前在中小板反转效应和动量效应的研究领域还存在较大的空白。本文试图探究中小板市场的股票价格运动规律,考证是否存在市场异象,以期为金融理论的进一步发展提供参考。第二、从其社会意义上来说,本文试图找到反转效应出现的原因,可以为证监会等政府机构提供参考,为政策的制定提供依据,促进资本市场的成熟,使其变得更加有效,更好的发挥其作为直接融资渠道的作用。第三、从投资者的角度来看,掌握了市场的规律,有助于他们的投资决策,并且获得超过市场平均水平的回报。当然如果存在无风险套利空间,从长期来看,是会随着投资者的充分竞争而消失的。本文的创新点主要表现在：第一、选择中小板市场作为研究对象的研究很少,只有王之飞(2011)对中小板进行了分析,但是由于当时中小板成立时间尚短,能够成为有效样本股的只有近五十只,从现在来看不具有代表性。本文选择的时间点很合适,有239只有效样本股,近四年半的完整牛熊市周期,能够全面的反映中小板的市场特征。第二、并未使用传统的CAR(累积超额收益率法)而是使用BH(买入并持有法)对收益率进行计算。BH(买入并持有法)方法,能更准确的反映股票的收益率,消除CAR(累积超额收益率法)可能存在的问题。第三、论文对中小板反转超额利润进行分解以探究利润的来源,在目前有关中小板市场动量效应、反转效应的研究当中尚属首例。并且论文发现了单因素定价模型分解的利润结果与存在性实证研究相悖的现象,推测了可能的原因。并经过进一步的研究发现,单因素定价分解模型本身存在的问题导致了该现象的产生。最后采用C-K恒等式对期望利润进行分解,得到了可靠的实证结果。论文主要分为以下几个部分：第一章、绪论介绍论文的选题背景、研究意义、研究内容及行文结构。第二章、文献综述主要从反转效应、动量效应的存在性和相应策略超额利润的来源这两个角度着手,搜集、整理了比较有代表性的国外国内的相应研究成果,并予以总结评述。第三章、反转效应、动量效应存在性的实证研究本文从中小板近700只股票中选择了在我设定的牛熊市周期有着较为完整交易记录的239只股票作为样本对象。首先从整体角度检验了反转效应、动量效应的存在性。其次,从牛熊市不同的周期阶段分别对其进行检验。最后,依据换手率高低,将240只股票划分成高换手和低换手两个集合,分别考察两个独立集合的表现特征。为之后的反转效应来源的分析奠定了基础,并为投资者提出相关的建议。第四章、反转策略超额利润来源分析根据Jegadeesh和titman (1993)的研究方法,利用单因素模型分解期望利润,讨论了利润的来源,发现实证结果与第三章的实证结果相悖。第五章、实证结果矛盾的原因探究针对第四章中发现的问题,用WRSS法对反转效应的存在性进行了再一次的确认,进而排除了是组合构建方法的影响。紧接着,用比单因素定价模型更可靠的C-K恒等式分解模型对期望利润进行了分解,得到了更为接近事实的结果。第六章、结论及展望对论文进行总结并陈述自己的结论,讨论本文的有待改进之处,并为后来者提出相关建议。在完成了上述工作后,经研究得出以下结论：第一、经两种不同的组合构建方法验证,就整体时间区间而言,中国中小板市场确实存在显著的反转效应,而并没有观察到显著的动量效应第二、牛市环境下中国中小板市场,只有少数的策略组合表现出显著的反转效应；熊市环境下中国中小板市场可以观察到显著的反转效应。第三、高换手率股票组表现较明显的是反转效应,而低换手率组同时存在显著的动量效应和反转效应。第四、中国中小板市场并非一个有效市场,中国中小板市场的反转收益主要来源于时间序列的可预测性。虽然我国中小板市场具有较大的横截面方差,对反转利润有削弱的作用,但时间序列的相关性依然占主导作用。就未来进一步研究的方向而言,可以在模型的选择上考虑使用多因素定价模型,但是难点在于多因素模型中的各个因素的选择十分讲究,选择不当会与现实差距很大；另外在数据处理时应考虑到自相关性的影响,可以采用Newey-west的方法进行调整。
[Abstract]:Since Fama proposed effective market theory in 60s, academic circles have been arguing about the existence of effective markets. Effective market theory holds that in the market, investors who have a lot of rationality and want to maximize their interests enjoy the same information, and their way of reading information is the same, the current stock Ticket prices have fully reflected all available information in the market. This theory prevailed in 1970s and was once honored by the academic world as the Bible.
But after 80s, some scholars have found that some stock movement phenomena are difficult to use effective market theory to explain.1985 Debont and Thaler (1985) put forward reversal effect. It refers to the ranking of a series of stock returns at a certain time point, which is called the loser, and the worst performance is called the loser. The combination, and the dozens of best performing groups called winners, then examined their performance in the middle and long term (2 to 5 years). They found a reversal, and the yield of the loser combination was better than the winner. The phenomenon was called the reversal effect. The first momentum effect proposed by Jegadeesh and titamn (1993) is also known as the inertia effect. In the next short term (3 to 12 months), the profit rate of the winner is higher than that of the losers, and the excess returns can be obtained by buying a winner portfolio, selling the loser portfolio and building a zero portfolio, and this investment strategy is called the momentum strategy.
For the purpose of exploring the laws of the reverse and momentum effects of the country, the author chooses the medium and small plate Market in China as its research object. Since the small and medium-sized board market has been established for nearly ten years, the number of stocks has increased to more than seven hundred, and the amount of the sample has been accumulated enough for the study of the short-term reversal effect. The main points are as follows:
First, from the academic point of view, there is still a big gap in the research field of small plate reversal effect and momentum effect. This paper tries to explore the law of stock price movement in the small and medium market market, and examines whether there is a market vision, in order to provide reference for the further development of the financial theory.
Second, from its social significance, this article tries to find the reasons for the reversal effect, which can provide reference for the government agencies such as the SFC, provide the basis for the formulation of the policy, promote the maturity of the capital market, make it more effective, and better play its role as a direct financing channel.
Third, from the perspective of investors, mastering the law of the market helps them to make their investment decisions and gains more than the market average. Of course, if there is a risk free arbitrage space, in the long run, it will disappear with the full competition of the investors.
The main innovation points of this paper are as follows:
First, the selection of small and medium board market as the research object is very few. Only Wang Zhifei (2011) has analyzed the small and medium board, but because the time of the small and medium-sized board is short, only nearly fifty can become effective sample stock, and it is not representative from now on. The time point of this paper is very suitable and there are 239 effective sample stocks. Nearly four and a half years, the complete bull bear market cycle can fully reflect the market characteristics of small and medium-sized boards.
Second, instead of using the traditional CAR (cumulative excess return rate method), we use the BH (buy and hold method) to calculate the rate of return (buy and hold method), which can more accurately reflect the stock returns and eliminate the possible problems of the CAR (cumulative excess rate of return).
Third, the paper decomposes the excess profit of small and medium plate reversals to explore the source of profit. It is the first one in the study of momentum effect and reversal effect in the medium and small market market. And the paper finds out the phenomenon that the profit result of the single factor pricing model is contrary to the existence empirical research, and the possible reasons are speculated. After further research, it is found that the problem of the single factor pricing decomposition model itself leads to the emergence of this phenomenon. Finally, the C-K identity is used to decompose the expected profit, and a reliable empirical result is obtained.
The thesis is divided into the following parts:
The first chapter introduces the background, significance, content and structure of the thesis.
The second chapter, the literature review mainly from the two angles of the reversal effect, the existence of momentum effect and the source of the corresponding strategic excess profit, collects and collects the relative research results of the representative foreign and domestic, and gives a summary and review.
The third chapter, reversal effect, the existence of momentum effect, this paper selects 239 stocks which have a complete transaction record in the period of the bull bear market which I set in 700 stocks as sample. First, we test the reversal effect and the existence of momentum effect from the whole point of view. Secondly, from the different weeks of the bull bear market Finally, according to the turnover rate, the 240 stocks are divided into two sets of high turnover and low turnover, respectively, to investigate the performance characteristics of the two independent sets, which lays the foundation for the analysis of the source of the reversal effect and puts forward the suggestions for the investors.
The fourth chapter, based on the research method of Jegadeesh and Titman (1993), a single factor model is used to decompose the expected profit, and the source of profit is discussed. It is found that the empirical results are contrary to the empirical results of the third chapter.
The fifth chapter, the reasons for the contradiction of the empirical results are explored in the fourth chapter, and the existence of the reversal effect is confirmed again by the WRSS method, and then the effect of the combined construction method is excluded. Then, the expected profit is decomposed by the C-K identity decomposition model which is more reliable than the single factor pricing model. The result is closer to the fact.
The sixth chapter, conclusion and prospect, summarizes the paper and states its own conclusions, discusses the improvements that need to be made in this paper, and puts forward some suggestions for the later.
After completing the above work, the following conclusions are drawn from the study.
First, through two different combination construction methods, the small plate Market in China does have a significant reversal effect in terms of the overall time interval, and the significant momentum effect is not observed.
Second, only a few strategic combinations show significant reversal effect in the medium and small plate Market in the environment of the bull market, and a significant reversal effect can be observed in the medium and small plate market of China under the bear market.
Third, the high turnover rate group's performance is more obvious than the reverse effect, while the low turnover group has significant momentum effect and reversal effect.
Fourth, the small and medium plate Market in China is not an effective market. The reverse return of China's small and medium-sized board market is mainly due to the predictability of the time series. Although the small and medium board market in China has a large cross section variance, it has a weakening effect on the reverse profit, but the correlation of the time series still occupies the leading role.
As far as future research is concerned, the multi factor pricing model can be considered in the selection of the model, but the difficulty is that the selection of various factors in the multi factor model is very particular, and the improper selection will be very large. In addition, the influence of autocorrelation should be taken into account in the process of data processing, and the Newey-west's side can be used. The law is adjusted.

【学位授予单位】：西南财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224

【参考文献】