中国中小板动量效应及反转效应实证研究

发布时间：2018-01-20 02:48

本文关键词： 动量效应反转效应中小企业板市场　出处：《西南财经大学》2014年硕士论文　论文类型：学位论文

【摘要】：有效市场假说(Efficient Market Hypothesis)被认为是现代金融学的基础,它的提出为现代金融学中很多重要理论与模型,例如资本资产定价模型的构建,提供了理论的基础。有效市场假说认为,在投资者均为理性经纪人的前提下,市场中的所有信息都能被投资者获得,并且这些信息能够迅速地,准确地反应在证券价格上。证券的价格是服从随机游走的,因此无论投资者怎么样选择自己的投资策略,都不能够获得超额利润,所以被动投资是最佳的投资策略。随后Fama按市场有效性的程度,将有效市场划分为了三类：弱有效市场,半强有效市场,强有效市场；并将市场上的信息划分为了：历史信息,公开信息和未公开的信息。弱有效市场的股票价格包含了所有的历史信息,投资者无法通过对历史数据的研究来获得超额的收益；半强有效市场指的是市场上的价格不但包含了弱有效市场中所有的历史信息,也包含了与该公司相关的所有的公开信息,投资者也无法通过这些信息去获取超额的收益；强有效市场是指市场的价格以及包含了所有的信息,包括历史信息,公开的信息以及未公开的信息,投资者也无法获取超额收益。然而随着研究的不断发展,学者们发现真实的市场确是与有效市场假说有着很大的差距。首先,投资者并不总是理性的经济人,噪声交易者在市场中的比例相当大；其次,市场中投资者的决策是具有惯性的,所产生的非理性的交易并不能因为大量的随机交易而被消除。且在实证中,学者们发现很多的市场异象,这些市场异象均不能通过有效市场假说和传统的资本资产定价模型来解释,反而对有效市场假说产生了冲击。在这诸多的市场异象中,本文所研究的动量效应与反转效应便是典型的代表.De BondtThaler(1985)在对纽约交易所1926-1982年的交易数据进行研究时,首次验证了美国股市上的反转效应的存在；动量效应则是由JagadeeshTitman(1993)对纽交所和美国证券交易所1965-1989的交易数据研究时验证——在美国的证券市场存在动量效应。在此基础上,后来的学者又对于其他国家市场上的动量效应及反转效应进行了验证以及在交易量、市场周期、国家地区因素等变量下的稳健性检验。并尝试运用定价模型和行为金融学的理论来对动量效应及反转效应的成因做出解释。在行为金融模型中,一般认为,动量效应是由于反应不足引起的,因为投资者对市场信息的反应滞后,所以赢者组合的股价是低于内在价格的；而反转效应是由于反应过度引起的,由于市场上过度的反应,使得股价高于内在价格,而随着时间的推进,股价最终会回归内在价值,因而形成反转效应。中国学者关于国内证券市场上的动量效应及反转效应的研究时,普遍认为在中国的证券市场上的动量效应存在于短期,反转效应存在于长期。在对中国证券市场上的动量效应及反转效应进行研究时,形成期与持有期一般采用周或者月进行划分。而在对于动量效应的成因上,有学者从市场结构、市场周期、交易量、个股特征等多方面进行了解释,但是目前还未有对于中国证券市场的动量效应及反转效应成因的统一的解释。国内研究者对于证券市场的动量效应及反转效应的研究,大多数都集中与主板市场,而对于本文所研究的中小板市场探讨较少。而笔者认为中小板市场经过将近10年的发展,已经成为了中国证券市场上一个十分重要的组成部分,同时由于中小板市场本身的特征是非常明显的,与主板市场有很大不同,所以对中小板市场的独立研究,就显得特别重要。并且在中小板这10年的发展中,已经为研究提供了充分的数据使得研究可行,因此本文选取中国中小企业板为研究对象,对其进行研究动量效应及反转效应的相关验证,争取能够为国内相关的研究提供更多的信息。本文的主要研究方法为,首先借鉴JegadeeshTitman(1993)的研究方法,对中国中小板市场2005年12月9日至2013年12月31日间符合条件的的中小板股票,共计161支进行验证。本文设定的形成期(J)为：1周,3周,6周,9周,12周；持有期(K)为：1周,3周,6周,9周,12周。通过交叉组合共形成25种组合策略。同时为增强数据的有效性和意义,本文采取重叠抽样(overlapping periods)的方法。具体为在每个t周开始时,股票将根据过去J期的的超额收益率进行降序排序,根据排序,前十分位为赢者组合,后十分位为输者组合,买入赢者组合,卖出输者组合,持有这一头寸K期构建一组动量即期投资策略组合(J,O,K)。同时,由于JegadeeshLehammn(1990)在研究中之初,股票的买卖价差,领先—滞后效应,价格压力等会影响到股票的收益率情况。因此,本文还将构建持有期较形成期滞后一周的,滞后一周动量效应组合(J,1,K)。并且在此基础上对不同市场周期下的动量组合的表现作出了检验。最后,本文根据相关实证结果,运用改进后的行为金融学的HS模型对结果做出解释。本文的实证结果发现,我国中小板市场存在着明显的动量效应。在极短期内(形成期与持有期1周之内)有着反转效应的存在,但是随着期限的延长,反转效应消失,动量效应变得明显。动量效应的来源主要是来自于组合中的赢者组合,说明赢者组合在中小板市场上的动量效应中起着主导的作用。在结果中,还发现,动量组合的收益与显著性随着持有期的延长而增长,但是达到9-12周,增速会明显的放缓。而通过验证滞后一周之后的动量组合,我们发现中小板市场上的动量组合更为显著,极短期内的反转效应也消失了。这在一方面证实了我国中小板市场上动量效应的存在以及在许多研究中提到的滞后一期的动量组合会表现更好,另一方面也说明了我国中小板市场的反转周期极短。在对中小板市场做分段检验时,我们发现在不同的市场周期中,动量组合的表现差异很大。在整体的牛市周期中,动量组合表现出了显著的动量效应；而在分段的牛市周期中,动量组合出现了动量效应、反转效应、动量效应交替出现的特征。在整体的熊市阶段,动量组合表现出了显著的反转效应；在分段的熊市周期中,动量组合在一阶段熊市周期中出现了反转效应,在二阶段的熊市周期中则不明显。这说明了我国中小板市场上的动量效应与反转效应与市场周期有着紧密的联系。随后本文使用了改进之后的HS模型对本文的研究结果做出了相应的解释,在标准的HS模型中加入了新的投资者类型“特殊动量交易者”,讨论了不同的交易者在各个阶段的交易行为会怎样作用于动量效应。结果发现,在极短期内的反转效应主要是由于特殊动量交易者打压股价,以及普通动量交易者对信息的反应过度；在极短期之后的动量效应主要是由于信息观察者对信息的反应不足,以及普通动量交易者、特殊动量交易者采取的动量交易策略；在持有期达到9-12周,增速明显放缓,动量效应依然存在时但已经开始减弱时,是因为特殊交易者的虚假造势,秘密出货以及普通交易者的盲目跟进所造成的。随后分析了在模型中出现的三类不同的投资者在我国证券市场上的地位。根据本文的研究结果及解释,笔者在文中最后给出了相关的政策建议,希望在在中小板市场中保护中小投资者的利益及严厉打击市场中人为操纵的行为。本文的研究结果也有着一定的不足,一方面是前提假定与现实的不完全符合,本文的结果是在假定中小板市场中交易成本为零,且证券买卖数量细分的前提下得出的,而这一点在现实中很难满足；另一方面,文中对于动量效应的成因的分析不够深入,得出的解释还需要进一步地深入。从整体上看,本文的研究对于中国整个市场有效性的探讨是有意义的及作出了相关的补充的。希望本文对于中小板的研究能够对主板市场的相关研究起到一定的补充作用,对创业板的相关研究有着一定的启发及借鉴的作用。
[Abstract]:Efficient market hypothesis (Efficient Market Hypothesis) is considered the foundation of modern finance, it is proposed for modern finance in many important theories and models, such as the construction of the capital asset pricing model, and provides a theoretical basis. The efficient market hypothesis holds that investors are in the premise of rational agents, all the information in the market can be obtained and these investors, information can quickly, accurately reflect in the stock price. The stock price is random walk, so no matter how investors choose investment strategy of their own, are not able to obtain excess profits, so the passive investment is the best investment strategy. Then according to the Fama Market the degree of effectiveness, the market will be divided into three categories: weak efficient market, semi strong efficient market and strong efficient market; and market information is divided into: history information, Public information and public information. The weak effective market stock price contains all the historical information, investors can not through the study of historical data to obtain excess returns; semi strong efficient market refers to the market price not only contains all the historical information of the weak efficient market, also includes the relevant public information and all of the company's investors, nor through the information to obtain excess returns; strong effective market refers to a market price and contains all the information, including historical information, public information and the information is not public, investors can obtain excess returns.
However, with the development of research, scholars have found that the real market is indeed there is a great gap with the efficient market hypothesis. First, investors are not always rational, the proportion of noise traders in the market is quite large; secondly, investors in the market are decision-making inertia, generated by the irrational trading and cannot be eliminated because of a large number of random transactions. And in the empirical research, scholars have found that many market anomalies, the market anomalies were not explained by the efficient market hypothesis and the traditional capital asset pricing model, but the impact on the efficient market hypothesis.
In this many market anomalies, the momentum effect and reversal effect is the typical representative of.De BondtThaler (1985) on the research of the New York stock exchange 1926-1982 years of trading data, first verify the reversal effect on the U.S. stock market; momentum effect is by JagadeeshTitman (1993) on the NYSE the American Stock Exchange and transaction data of the 1965-1989 verification of the securities markets in the United States had momentum. On this basis, later scholars and other countries on the market for the momentum effect and reversal effect was verified as well as in trading volume, market cycle, test the robustness of variable factors such as regional countries under and. Try to use the theory of pricing model and behavioral finance to the causes of momentum effect and reversal effect to explain. In the behavioral model, is generally believed that the momentum effect Is due to the reaction caused by insufficient, because investors reacted to the market information lag, so the winners share price is lower than the intrinsic price; while the reversal effect is due to the reaction caused by the excessive, because the market overreaction, the price is higher than the internal price, but as time progresses, the stock price will eventually return to the intrinsic value. Thus forming a reverse effect.
Study on momentum effect China scholars on the domestic stock market and the reversal effect, momentum effect in China generally believe that the securities market exists in the short term, the reversal effect exists in the long term. Study on momentum effect of Chinese on the securities market and the reversal effect, the formation period and holding period by the week or month division. While in the causes of momentum effect, some scholars from the market structure, market cycle, trading volume, stock characteristics and other aspects of interpretation, but there is no unified Chinese for momentum effect in the securities market and the reversal effect of explanations of the causes.
Study on the momentum effect of stock market and the reversal effect of domestic researchers, most of them are concentrated with the motherboard market, but for the small and medium-sized market research in this paper is less discussed. The author believes that after nearly 10 years of development of small and medium-sized market, has become the China securities market is an important part of, at the same time due to the characteristics of small and medium-sized market itself is very obvious, is very different from the motherboard market, so the independent research of small and medium-sized board market, it is particularly important. In the small board and the 10 years of development, the research has provided sufficient data makes it feasible, therefore this paper selects Chinese SME board as the research object, related research and verification on the momentum effect contrarian effect, and strive to be able to provide more information for the domestic related research.
The main methods for this study, the first reference JegadeeshTitman (1993) the research methods of small and medium-sized board stock market from December 9, 2005 to December 2013 China small plates 31 day in line with the conditions, a total of 161 validated. This paper set the formation stage (J) for 1 weeks, 3 weeks, 6 weeks, 9 weeks, 12 week; holding period (K) for 1 weeks, 3 weeks, 6 weeks, 9 weeks, 12 weeks. The cross combinations were formed 25 kinds of combination strategies. At the same time in order to enhance the data validity and significance, this paper adopts overlapping sampling (overlapping periods) method. The t at the beginning of each week. The stock will be the basis of past excess returns J phase rate in descending order, according to the order, is the former for the winners, is for losers, buying winners, selling losers, holding this position K to build a set of momentum investment strategy at sight (J, O, K) at the same time, due to the JegadeeshL Ehammn (1990) in the study of the early trading in the stock price, lead lag effect, price pressure will affect the stock return. Therefore, this paper will construct the holding period is the formation period lag a week, a week lag momentum portfolio (J, 1, K). The combination of momentum on the performance of different market cycles and on the basis of making test. Finally, according to the empirical results, using the HS model of behavioral finance improved the interpretation of the results.
The empirical results show that China's small and medium-sized market exists significant momentum effect. In the very short term (the formation period and holding period of 1 weeks) has a reverse effect, but with the extension of time, the reversal effect disappears, the momentum effect becomes obvious. The source of the momentum effect is mainly derived from the combination of the winners, that plays a leading role in the small and medium-sized board momentum winners on the market. Also found in the results, and the significant revenue growth momentum as the holding period extended, but at 9-12 weeks, the growth rate will slow down significantly. The lag momentum a week after verification we found that the combination of small and medium-sized board market momentum is more obvious, the reversal effect of the very short term, also disappeared. This confirms the momentum of China's small and medium-sized board on the market and in many studies in hand The momentum combination of the lagging one will perform better. On the other hand, it also shows that the reverse cycle of the small and medium plate markets in China is very short.
In subsection test of small and medium-sized board market, we find that in different market cycles, differences of momentum portfolios greatly. In the overall bull market cycle, the momentum portfolios showed significant momentum effect; while in the segmented bull market cycle, the combination of momentum momentum effect and reversal effect, momentum effect the alternation of characteristics. In the phase of the bear market overall, momentum showed a significant reversal effect; in the section of the bear market cycle, the momentum reversal effect in a bear market cycle stage, in the two phase of the Xiong Shizhou period was not obvious. This shows that the momentum effect in China's SME board on the market and the reversal effect and market cycles are closely linked.
Then this paper uses the HS model to improve the results of this paper is to make the explanation, adding a new type of investors "special momentum traders" in the standard HS model, discusses the trading behavior of different traders in various stages of what will effect on the momentum effect. The results showed that the inversion effect in the very short term the main is due to the special momentum traders price pressure, and ordinary momentum traders react to information excessive; momentum effect in a very short after the information is mainly due to the observer response to the lack of information, as well as ordinary momentum traders, momentum momentum traders take special trading; in the holding period up to 9-12 weeks, growth is slowing down the momentum effect still exists, but has been weakened, because false campaign special traders, secret shipments and ordinary traders Blindly follow caused. Then analyze the status of three different types of investors in the model in Chinese stock market. According to the results of this study and interpretation, the author finally gives relevant policy suggestions, in the hope that the protection of the interests of small investors in small and medium-sized board market and crack down on the market in the manipulation behavior.
The results of this study also have some problems, one is the premise and the reality of the not entirely consistent with the result of this paper is on the assumption that the small and medium-sized board market transaction cost is zero, and the number of segments of the securities trading under the premise, which in reality is very difficult to meet; on the other hand, analysis of the causes for the momentum effect in this paper is not deep enough, it also needs further explanation.
On the whole, this research is meaningful and has made the relevant supplement to explore the effectiveness of China whole market. I hope this research for small and medium-sized plate related research on the main market play a complementary role, has a certain inspiration and reference of the related study on GEM.

【学位授予单位】：西南财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F832.51

【参考文献】