A股股票估值和基本面要素的实证分析
发布时间：2018-07-16 18:10
【摘要】：国内证券市场已经发展近20多年时间。市场上出现了许多新的变化,如股票的规模不断扩大,市场机制不断完善,不断提高股票上市公司的信息披露,机构投资者的比例在增加。A股市场原先流行的坐庄方式已经越来越难以实行,百圆裤业的闹剧就充分说明了缺乏基本面,仅靠好看技术形态难以吸引了后续资金的介入。基本面要素已成为投资者在对A股股票进行估值时的重要考虑因素。通过对以往估值理论的回顾,本文发现过往的估值模型更多的关注基本面要素对于股价或超额收益的影响,对于估值的分析较少；以往论文给出的估值模型,很多只要稍加变换,就可以转化成估值和基本面要素的模型。此外,在实际股票投资中,估值是投资者实实在在考虑的事情,估值是否仅仅是投资者主观的判断,还是与基本面要素存在相关性?这是本文主要解决的问题。针对这个问题,本文采用了A股近三年来的上千家上市公司数据,构建估值与基本面要素之间的多因素模型,通过主成分分析的方式来考虑估值与基本面要素之间的关系。本文主要分成三个部分,在第一部分中,本文将对国内外主要股票估值理论以及方法进行梳理。资本市场中存在着各种不同的股票估值理论和方法,这些理论和方法已经形成严密的估值体系。按照这些模型的核心思想,本文可大体将这些模型分成贴现模型、相对估值方法、信息观理论和计量观估值模型等。在第二部分中,本文将基于国内外已有实证研究,结合A股市场的实际情况给出了本文的研究模型。本文认为因子分析可以很好满足本文的分析需要,在选择因变量指标时,本文选择了PE值作为因变量,这主要是由于PE值是实际估值中最为常用的指标,在选择了基本面指标时,本文选择了净利润增长率、波动率、行业PE等20个指标,并将这20个指标分成股本、现金流量、资产结构、交易活跃度等8类。本文分析的主要步骤如下表示：1、选择样本数据,对样本数据进行预处理,剔除那些异常值,对数据做无量纲化处理,调整数据中那些极端值；2、建立多因素模型,进行回归分析,对模型的系数进行分析；运用因子分析的方式,对数据做进一步分析。在第三部分中,本文分别以2011年、2012年、2013年这三年的数据作为本文分析的样本来源；本文选取了每年五月份的第五个交易日作为选取数据的节点。这主要是由于上市公司的年报更多是在3月和4月这两个月份出来,同时为了考虑市场对于消息的消化时间。本文最终的研究结论表明,基本面要素可以解释了40-50%的企业估值,资产结构、股本规模、经营能力、行情指标等基本面指标都对股票估值有影响；用模型给出的预测值和实际估值的差异给出低估组合和高估组合,然后对组合的未来股价表现进行检测,结果证明了模型是有效,也说明了基本面要素可以帮助投资者正确预测上市公司估值。
[Abstract]:The domestic securities market has developed for more than 20 years. There have been many new changes in the market, such as the constant expansion of the stock scale, the continuous improvement of the market mechanism, and the continuous improvement of the information disclosure of the listed companies. The proportion of institutional investors is on the rise. The A-share market has become more and more difficult to carry out, the farce of the Baiyuan pants industry fully illustrates the lack of fundamentals, relying solely on good-looking technology form to attract follow-up funds to intervene. Fundamental factors have become an important consideration for investors in the valuation of A-shares. By reviewing the previous valuation theory, this paper finds that the previous valuation models pay more attention to the influence of fundamental factors on stock price or excess return, and less to the analysis of valuation. Many can be transformed into models of valuation and fundamental elements with a little transformation. In addition, in the actual stock investment, valuation is a real consideration for investors. Is valuation only a subjective judgment of investors, or is it related to fundamental factors? This is the main problem solved in this paper. In order to solve this problem, this paper uses the data of thousands of A-share listed companies in the past three years, constructs a multi-factor model between valuation and fundamental factors, and considers the relationship between valuation and fundamental factors through principal component analysis. This paper is divided into three parts. In the first part, this paper will sort out the main stock valuation theories and methods at home and abroad. There are a variety of stock valuation theories and methods in the capital market, and these theories and methods have formed a rigorous valuation system. According to the core ideas of these models, these models can be divided into discount model, relative valuation method, information theory and econometric valuation model. In the second part, based on the domestic and foreign empirical research, combined with the actual situation of A-share market, this paper gives the research model. This paper thinks that factor analysis can well meet the needs of the analysis in this paper. In selecting the dependent variable index, this paper chooses PE value as the dependent variable, which is mainly because PE value is the most commonly used index in the actual valuation. This paper selects 20 indexes, such as net profit growth rate, volatility rate, industry PE and so on, and divides them into eight categories: equity, cash flow, asset structure, trading activity and so on. The main steps of this paper are as follows: 1, select the sample data, preprocess the sample data, eliminate the outliers, do dimensionless processing on the data, adjust the extreme values in the data, and establish a multifactor model. Regression analysis is carried out to analyze the coefficients of the model, and further analysis of the data is made by using the method of factor analysis. In the third part, the data of 2011, 2012 and 2013 are taken as the sample source of this paper, and the fifth trading day of May is selected as the node of the data. This is mainly because the annual reports of listed companies are more likely to come out in March and April, as well as to consider the timing of the market's digestion of the news. The final conclusion of this paper shows that the fundamental factors can explain 40-50% of the enterprise valuation, asset structure, equity size, business ability, market index and other fundamental indicators have an impact on stock valuation; The difference between the predicted value and the actual valuation is used to give an undervalued combination and an overvalued combination, and then the future stock price performance of the portfolio is tested. The results show that the model is effective. It also shows that fundamental factors can help investors correctly predict the valuation of listed companies.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51
本文编号：2127211
[Abstract]:The domestic securities market has developed for more than 20 years. There have been many new changes in the market, such as the constant expansion of the stock scale, the continuous improvement of the market mechanism, and the continuous improvement of the information disclosure of the listed companies. The proportion of institutional investors is on the rise. The A-share market has become more and more difficult to carry out, the farce of the Baiyuan pants industry fully illustrates the lack of fundamentals, relying solely on good-looking technology form to attract follow-up funds to intervene. Fundamental factors have become an important consideration for investors in the valuation of A-shares. By reviewing the previous valuation theory, this paper finds that the previous valuation models pay more attention to the influence of fundamental factors on stock price or excess return, and less to the analysis of valuation. Many can be transformed into models of valuation and fundamental elements with a little transformation. In addition, in the actual stock investment, valuation is a real consideration for investors. Is valuation only a subjective judgment of investors, or is it related to fundamental factors? This is the main problem solved in this paper. In order to solve this problem, this paper uses the data of thousands of A-share listed companies in the past three years, constructs a multi-factor model between valuation and fundamental factors, and considers the relationship between valuation and fundamental factors through principal component analysis. This paper is divided into three parts. In the first part, this paper will sort out the main stock valuation theories and methods at home and abroad. There are a variety of stock valuation theories and methods in the capital market, and these theories and methods have formed a rigorous valuation system. According to the core ideas of these models, these models can be divided into discount model, relative valuation method, information theory and econometric valuation model. In the second part, based on the domestic and foreign empirical research, combined with the actual situation of A-share market, this paper gives the research model. This paper thinks that factor analysis can well meet the needs of the analysis in this paper. In selecting the dependent variable index, this paper chooses PE value as the dependent variable, which is mainly because PE value is the most commonly used index in the actual valuation. This paper selects 20 indexes, such as net profit growth rate, volatility rate, industry PE and so on, and divides them into eight categories: equity, cash flow, asset structure, trading activity and so on. The main steps of this paper are as follows: 1, select the sample data, preprocess the sample data, eliminate the outliers, do dimensionless processing on the data, adjust the extreme values in the data, and establish a multifactor model. Regression analysis is carried out to analyze the coefficients of the model, and further analysis of the data is made by using the method of factor analysis. In the third part, the data of 2011, 2012 and 2013 are taken as the sample source of this paper, and the fifth trading day of May is selected as the node of the data. This is mainly because the annual reports of listed companies are more likely to come out in March and April, as well as to consider the timing of the market's digestion of the news. The final conclusion of this paper shows that the fundamental factors can explain 40-50% of the enterprise valuation, asset structure, equity size, business ability, market index and other fundamental indicators have an impact on stock valuation; The difference between the predicted value and the actual valuation is used to give an undervalued combination and an overvalued combination, and then the future stock price performance of the portfolio is tested. The results show that the model is effective. It also shows that fundamental factors can help investors correctly predict the valuation of listed companies.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51
【参考文献】
相关期刊论文 前2条
1 王帅;陈石清;;PE新视角审视当前行业与估值[J];金融经济;2008年12期
2 杨成;谢鹏程;殷旅江;;钢铁业上市公司财务指标与股票价格定位实证研究[J];价值工程;2006年03期
，本文编号：2127211
本文链接：https://www.wllwen.com/jingjilunwen/guojijinrong/2127211.html
