基金经理投资风险行为机理分析和预测

发布时间：2018-01-04 02:40

本文关键词：基金经理投资风险行为机理分析和预测　出处：《东北大学》2014年硕士论文　论文类型：学位论文

【摘要】：近些年,开放式基金的规模逐年增加,已经逐渐成为中国金融市场重要组成部分。基金经理作为基金投资管理的主要负责人,其投资风险行为对于基金的长期业绩的稳定性具有重要影响作用。针对基金经理的投资风险行为的机理分析和预测,可以给投资者和监管者的决策提供参考依据,对于开放式基金的长期良性地发展具有重要意义。国内外大多数学者,采用基金业绩的波动,即基金业绩的标准差,作为基金经理投资风险的衡量标准。研究发现,基金经理会根据自己历史的相对排名对将来的投资风险进行调整。本文在此基础上,基于大量开放式基金运行数据和特征,分别采用了多元回归方法和神经元网络方法对基金经理投资风险行为进行了机理分析和预测。实证检验表明了方法的有效性和普遍性。首先,从职业生涯角度对基金经理投资风险行为机理进行分析,基金经理出于自利性的原因,在进行投资和做决策的出发点更多的是自己职业生涯考虑,以此为依据决定下期是否增大或减小投资风险,前期排名较低的基金经理面临更大的雇佣风险,在下期选择高风险策略,而前期排名较高的基金经理没有职业压力,为了维持劳动力市场上的声誉,在下期选择低风险的策略。其次,针对中国2011开放式股票型基金的面板数据,选取基金相对业绩排名等因素,利用多元线性回归方法对基金经理风险进行回归拟合,通过逐步回归方法求解出回归系数,建立回归预测模型,利用2012年开放式股票型基金运行数据进行了验证。最后,为了提高预测精度,分别采用了基于BP神经网络和RBF神经网络方法来预测基金经理风险行为。通过两种方法的对比分析显示,RBF神经网络预测模型预测准确性和稳定性方面都有显著性的提高。
[Abstract]:In recent years, open-end fund scale increased year by year, has gradually become an important part of the financial market. Chinese fund managers as the main person in charge of fund investment management, which plays a very important role in the stability of investment risk for long-term performance of the fund's behavior. The mechanism analysis and prediction of investment risk behavior for fund managers, can provide a reference for investors and regulators of the decision, the open-end fund's long-term healthy development has important significance. Most scholars at home and abroad, the fund performance fluctuations, namely fund performance standard deviation, as the fund manager's investment risk measure. The study found that the fund manager will according to the relative ranking of their own history to adjust the future the risk of investment. On this basis, a large number of open-end fund operation data and based on the features, using multiple back The regression method and neural network method on fund managers' investment risk behavior is analyzed and forecasted. The empirical test shows the effectiveness and universality of the method. First, analyze the mechanism of fund managers' investment risk behavior from the perspective of occupation career, the fund manager for reasons of self benefit, the starting point of investment and decision making the more of their own occupation career to consider, as a basis for deciding whether to increase or decrease the risk of investment, the lower ranked fund managers face a greater risk of employment, in the next higher risk strategy, the pre ranking fund managers have no higher occupation pressure, in order to maintain the reputation of the labor market, low risk the strategy in the next issue. Secondly, according to the panel data of 2011 China open stock fund, selected fund relative performance ranking factors, using multiple linearregression To the method of risk to the fund manager by regression fitting, stepwise regression method to calculate the regression coefficient, regression prediction model was verified by the 2012 open stock fund operation data. Finally, in order to improve the prediction accuracy, respectively based on BP neural network and RBF neural network method to predict the fund manager's risk behavior. By comparing the two methods of analysis shows that the prediction model prediction accuracy and stability has significantly improved RBF neural network.

【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP183;F832.51

【相似文献】