证券分析师对上市公司盈利预测的行为金融学解读

发布时间：2018-06-02 04:06

本文选题：证券分析师 + 盈利预测　；参考：《复旦大学》2013年硕士论文

【摘要】：证券分析师的主要工作是对上市流通证券的价值进行分析或做出预测,并给出相关投资建议。传统的金融学理论是建立在有效市场假设和资本资产定价理论体系之上,并且传统金融理论的一个十分重要前提假定便是人具有完全理性。如果按照传统的金融学理论,证券分析师毫无疑问应当属于“理性人”的范畴,即他们的预测行为将是完全基于他们对上市公司基本面的分析和对宏观、微观环境的判断而做出的,其中不会掺杂非理性的因素。显然这些假设是站不住脚的。本文的研究目的就在于研究中国的证券分析师在做出盈利预测行为的时候,是否受到了一些非理性因素的影响。如果是的话,这些无法通过传统金融学解释的部分,能否通过新兴的行为金融学理论得到部分解释,并且试图探讨这种带有非理性因素的预测行为的后果与影响。本文试图通过纵向比较以及横向比较这两种不同视角来看待证券分析师所表现出来的非理性部分。所谓纵向比较,是站在时间序列角度探讨证券分析师的预测行为是否前后相关联。通过实证检验可以发现,证券分析师本期的预测数值会受到其本人上一期预测结果的影响,并且若某位分析师前期做出的预测值低于了实际结果,那么本期该分析师更有可能做出一个过于乐观的预测值,反之亦然。这种非理性现象可以通过行为金融学中的“锚定效应”理论得到部分说明：证券分析师在做出预测之前,将初始值锚定在上一期的预测结果之上,并且这种结果直接影响到了其本期的预测,使得分析师更有可能出现反应过度的表现。所谓横向比较,是站在横截面角度探讨证券分析师的预测行为是否受到其他分析师的影响。本文分别以明星证券分析师和首先做出预测的分析师为参照对象,研究他们的预测结果是否会对其他分析师的预测产生影响。通过实证检验发现,明星分析师的预测结果对其他分析师是有较为显著影响的,即证券分析师有可能会做出和明星分析师相类似的预测。然而,分析师不太会参考首先做出预测的分析师的预测结果。可以用行为金融学中的“羊群效应”理论来解读这种非理性的现象,即可以认为明星分析师充当了“头羊”的角色,而其他分析师则扮演了“群羊”的角色。本文给出的结论如下：第一,证券分析师预测行为并非是完全理性的；第二,证券分析师会比较看重自己先前的预测结果否准确,并且这些信息会影响其当前的预测判断；第三；证券分析师会留意同行业其他人的预测行为,尤其是那些所谓的明星分析师。
[Abstract]:The main job of securities analysts is to analyze or forecast the value of listed securities and give relevant investment advice. The traditional financial theory is based on the efficient market hypothesis and the capital asset pricing theory system, and a very important premise of the traditional financial theory is that people have complete rationality. If according to the traditional financial theory, the securities analysts should undoubtedly belong to the category of "rational people", that is, their prediction behavior will be based entirely on their analysis of the fundamentals of listed companies and on the macro. Micro-environmental judgment, which does not mix with irrational factors. Obviously these assumptions are untenable. The purpose of this paper is to study whether Chinese securities analysts are influenced by some irrational factors when making earnings forecast. If so, can these parts, which cannot be explained by traditional finance, be partly explained by the emerging behavioral finance theory, and try to explore the consequences and effects of this kind of prediction behavior with irrational factors. This paper attempts to look at the irrational part of securities analysts from two different perspectives: vertical comparison and horizontal comparison. The so-called vertical comparison is to discuss whether the forecasting behavior of securities analysts is related to the former from the angle of time series. Through the empirical test, we can find that the forecast value of the current period of the securities analyst will be affected by their own forecast results in the previous period, and if the previous forecast value made by a certain analyst is lower than the actual result, So this issue the analyst is more likely to make an overly optimistic forecast, and vice versa. This irrational phenomenon can be partially explained by the "anchoring effect" theory in behavioral finance, in which securities analysts anchor their initial values above the previous prediction results before making predictions. And the results directly affect its forecast for the current period, making analysts more likely to overreact. The so-called horizontal comparison is a cross-sectional approach to see if the forecast behavior of securities analysts is influenced by other analysts. This paper uses star securities analysts and first forecasters as a reference to see if their results will have an impact on other analysts' forecasts. Empirical tests show that star analysts' forecasts have a significant impact on other analysts, that is, securities analysts are likely to make similar forecasts to star analysts. However, analysts are less likely to refer to the results of the first forecast. The theory of "herd effect" in behavioral finance can be used to interpret this irrational phenomenon, that is, star analysts play the role of "head sheep", while other analysts play the role of "herd sheep". The conclusions given in this paper are as follows: first, the behavior of securities analysts is not completely rational; second, the securities analysts will value the accuracy of their previous prediction results, and these information will affect their current prediction judgment. Third, securities analysts will pay attention to the expected behavior of the rest of the industry, especially those so-called star analysts.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F275

【相似文献】