基于公司价值和市场行为的股价波动模型比较、拓展及应用

发布时间：2018-04-30 05:20

本文选题：公司价值 + 市场行为　；参考：《福州大学》2014年硕士论文

【摘要】：我国股票市场成立至今,已经历了多次暴涨暴跌,从上证指数来看,低时探底至95.79点,高时达到6124.04点,股价的过度波动破坏了市场的价格发现与资本配置功能,加大了我国资本市场的投资风险,给我国经济发展造成了较大不良影响。在理性预期假设下,投资者会对公司的价值与风险进行预期。而在有效市场条件下,股价围绕其价值波动,公司价值的变动是股价产生波动的根本原因。投资者对风险的预期则会体现在市场行为上,行为金融理论尝试从投资者的异质特征和有限理性等假设出发,通过对投资者的心理活动进行分析,研究市场行为变化对股价波动的影响。本文在理性预期假设条件下,同时从公司价值和市场行为两个方面分别考虑其对股价波动的影响,并考虑这两方面对改进股价波动模型拟合程度与预测精度的可能帮助。一方面,本文从预期理论和有效市场理论出发,阐述了股价波动能够被预测的理论基础。对在有效市场条件下的基于现金流量、盈利和乘数的公司价值评估模型进行理论比较,发现基于股利折现、现金流量和异常盈利的价值评估模型本质上都是等价的；在对他们的应用比较中发现基于乘数的价值评估模型较适合应用于研究我国股价波动,而市盈率乘数是衡量我国上市公司价值的较为理想的指标。另一方面,受BSV.DHS和HS模型理论比较的启发,选择了市场行为中的投资者自信度、市场意见分歧和投资者情绪三个方面进行研究,并选取新增开户数、交易量和封闭式基金折价率作为相应的评价指标。随后,通过协整关系检验、建立误差修正模型以及脉冲响应和方差分解分析等方法,量化分析了公司价值及市场行为与股价波动之间的关系。结果表明,公司价值和市场的三方面行为与股指间皆存在协整的长期均衡关系；公司价值的提升有助于迅速拉升股价,且影响周期较长；股价本身、市场意见分歧和投资者自信度能在较大程度上对股价波动进行解释,投资者情绪对股价波动的影响显现较快。之后,考虑到股指较易受到前期波动和前期方差的影响以及波动集群性和异方差性等特征,选择ARMA-GARCH模型对股价的波动进行预测,同时在前述结论的基础上,为了全面考虑股价波动的长短期特征,对该波动模型从公司价值和市场行为方面进行了拓展。经实证检验证明该拓展模型的拟合度和预测精度较同类模型有显著的提高。最后,依据本文理论与实证分析结果为预防我国股价出现过度波动提出了合理化建议。
[Abstract]:Since the establishment of the stock market, China's stock market has experienced many spikes and plummets.From the Shanghai Stock Exchange index, the low market bottom reached 95.79 points and the high stock market reached 6124.04 points. The excessive fluctuation of the stock price undermined the market's function of price discovery and capital allocation. It increases the investment risk of our country's capital market, and brings great bad influence to our country's economic development. On the assumption of rational expectation, investors will expect the value and risk of the company. Under the efficient market conditions, the stock price fluctuates around its value, and the change of the company value is the fundamental reason for the fluctuation of the stock price. Investors' expectation of risk will be reflected in market behavior. Behavioral finance theory tries to analyze the psychological activities of investors from the assumption of investors' heterogeneity and limited rationality. To study the effect of market behavior change on stock price fluctuation. In this paper, under the assumption of rational expectation, we consider the influence of company value and market behavior on stock price volatility, and consider the possible help of these two aspects to improve the fitting degree and prediction accuracy of stock price volatility model. On the one hand, based on expectation theory and efficient market theory, this paper expounds the theoretical basis that stock price volatility can be predicted. Through the theoretical comparison of the valuation model based on cash flow, profit and multiplier under the efficient market conditions, it is found that the valuation model based on dividend discount, cash flow and abnormal profit is essentially equivalent; In the comparison of their applications, it is found that the multiplier based valuation model is more suitable for studying the volatility of stock price in China, and the price-earnings multiplier is an ideal index to measure the value of listed companies in China. On the other hand, inspired by the comparison of BSV.DHS and HS model theory, we choose three aspects of investor confidence, market opinion difference and investor sentiment in market behavior, and select the number of new account opening. Trading volume and closed-end fund discount rate as the corresponding evaluation indicators. Then, through cointegration test, error correction model, impulse response and variance decomposition analysis are established to quantitatively analyze the relationship between company value, market behavior and stock price volatility. The results show that there is a long-term equilibrium relationship between corporate value and market behavior and stock index; the promotion of corporate value contributes to the rapid rise of stock price, and has a long period of influence; the stock price itself, Market disagreement and investor confidence can explain stock price volatility to a greater extent, and investor sentiment has a faster impact on stock price volatility. Then, considering that the stock index is vulnerable to the influence of early volatility and early variance, as well as volatility cluster and heteroscedasticity, ARMA-GARCH model is selected to predict the volatility of stock price. At the same time, on the basis of the above conclusions, In order to comprehensively consider the short and long term characteristics of stock price volatility, the volatility model is extended from the aspects of firm value and market behavior. It is proved by empirical test that the fitting degree and prediction accuracy of the extended model are significantly higher than that of the similar models. Finally, according to the results of theoretical and empirical analysis, some reasonable suggestions are put forward to prevent excessive fluctuation of stock price in China.
【学位授予单位】：福州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【相似文献】