社会网络中的管理者自我修正过度自信的实证研究

发布时间：2018-04-01 20:10

本文选题：管理者过度自信　切入点：自利归因度量法　出处：《沈阳工业大学》2017年硕士论文

【摘要】：心理学家研究表明,大多数各行各业的人们往往高估所拥有的知识和判断信息的准确性。因此,他们都具有过度自信的认知偏差。管理者过度自信是指管理者在做投、融资等公司战略性决策时会过高的估计自身能力和知识,高估成功的概率和低估风险水平,从而产生认知偏差。管理者的过度自信对企业投资决策、企业融资行为和企业并购活动都会产生不利影响。因此,如何修正管理者过度自信认知偏差,是管理实践中急需解决的问题。本文首先对国内外的已有文献研究进行了回顾总结,找出影响管理者过度自信的原因,发现管理者的学习行为能较好地修正过度自信心理偏差。再依据社会网络相关理论,发现处于社会网络中的管理者可以获得更多的学习机会。在此理论基础之上,提出本文的研究假设,管理者网络位置与管理者过度自信程度成负相关关系以及并购历史在两者关系之间具有调节作用。然后,进行变量设计与模型构建。本文选取2014年深沪A股上市公司作为样本,运用UCINET软件,利用ACCESS数据库,构建管理者社会网络。以管理者过度自信指标为被解释变量,以管理者社会网络位置指标为解释变量,以并购历史为调节变量,运用逻辑回归方法研究处于不同网络位置的管理者与其过度自信程度的关系,并研究并购历史对管理者网络位置与管理者过度自信两者之间关系的影响。最后,选用另一种度量管理者过度自信程度的方法,进一步检验管理者网络位置、并购历史与管理者过度自信之间的关系。研究结果发现:管理者网络位置与管理者过度自信程度显著负相关,揭示处于网络中心位置的管理者,其过度自信程度越低。管理者网络位置与并购历史的交互项与管理者过度自信成显著正相关关系,说明并购历史弱化了管理者网络位置与管理者过度自信的负相关关系。这表明位于社会网络中的管理者通过获得学习机会,修正了管理者过度自信认知偏差,扩展了社会网络研究新的应用领域。
[Abstract]:Psychologists have shown that people in most industries tend to overestimate the accuracy of their knowledge and judgment information. As a result, they all have cognitive biases of overconfidence. Companies such as financing will overestimate their own ability and knowledge, overestimate the probability of success and underestimate the level of risk, which results in cognitive bias. Therefore, how to correct the cognitive bias of managers' overconfidence, It is an urgent problem to solve in management practice. Firstly, this paper reviews and summarizes the existing literatures at home and abroad, and finds out the reasons that affect managers' overconfidence. It is found that the learning behavior of managers can correct the psychological bias of overconfidence. Then, according to the theory of social network, it is found that managers in social networks can get more learning opportunities. This paper proposes the hypothesis that the location of manager network has a negative correlation with the degree of manager overconfidence and the history of M & A has a moderating effect between the two. Using UCINET software and ACCESS database, the social network of managers is constructed. The overconfidence index of managers is taken as the explanatory variable, which is based on the index of overconfidence of managers, which is selected as the sample of the listed companies in Shenzhen and Shanghai in 2014, and the social network of managers is constructed by using ACCESS database. Taking the social network location index as the explanatory variable and the M & A history as the adjusting variable, the relationship between the managers in different network locations and their degree of overconfidence is studied by using the logical regression method. The relationship between managers' network location and managers' overconfidence is studied. Finally, another method to measure the degree of manager's overconfidence is used to further test the managers' network location. The relationship between the history of M & A and the overconfidence of managers. The results show that the network location of managers is negatively related to the degree of overconfidence of managers, which reveals that the managers are in the center of the network. The lower the degree of overconfidence is, the more significant positive correlation exists between the interaction between manager's network location and M & A history and manager's overconfidence. It shows that M & A history weakens the negative correlation between managers' network location and managers' overconfidence, which indicates that managers in social networks can correct the cognitive bias of managers' overconfidence by gaining learning opportunities. The new application field of social network research is expanded.
【学位授予单位】：沈阳工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F275;F832.51

【参考文献】