基于家庭消费资产定价模型的实证分析

发布时间：2018-03-27 09:28

本文选题：家庭生命周期理论　切入点：CCAPM　出处：《厦门大学》2014年硕士论文

【摘要】：基于消费的资产定价模型(CCAPM)理论上非常完美,然而实证检验显示该模型估计的风险厌恶系数远远高于合理值(3),这也是造成股权溢价之谜的主要原因。笔者认为并不是所有的消费者均参与股票市场,因此结合人口结构变化和家庭生命周期理论对消费者进行分类,构造新的代理人。通过实证分析,寻求真正决定市场资产价格的代理人和较为合理的风险厌恶系数。根据家庭生命周期理论,笔者构造个人代理人和家庭代理人,对基于长期消费风险的资产定价模型进行实证分析。本文选取1960年至2011年美国居民消费数据和Fama French25个股票组合收益率为样本。构造以年龄为分组标准的个人代理人和以平均年龄为分类依据的家庭代理人,进行CCAPM的实证分析。以个人代理人的消费数据为样本进行实证分析,笔者发现45岁以上的个人代理人消费数据能够更好的解释模型,并且计算出来的风险厌恶系数更加符合实际的风险厌恶系数。以家庭代理人的消费数据为样本进行实证分析,笔者发现股票市场价格的决定者是家庭平均年龄较大的家庭,计算出来的风险厌恶系数比以总体消费风险为样本的计算结果小,并且模型的拟合度也远远高于其他家庭代理人。由于本文人口结构数据的缺乏,笔者进行了CGF (Comsuption Growth Factor-mimicking Portfolios)组合的构造预测消费数据,并进行了实证分析,实证结果进一步证实了上述的结论。本文的研究结果表明,以家庭消费为研究对象比以个人为研究对象能更好的解释资产定价模型。在市场上,真正决定资产价格的是没有抚养和教育子女义务等负担的家庭。同样本文也再次验证了长期消费风险能够更好的估计资产的价格,验证了基于长期消费风险的资产定价模型的准确性。
[Abstract]:The consumption-based asset pricing model, CCAPM-based, is theoretically perfect. However, the empirical test shows that the estimated risk aversion coefficient of the model is much higher than the reasonable value, which is also the main reason for the mystery of equity premium. I think that not all consumers participate in the stock market. Therefore, based on the theory of population structure change and family life cycle, we classify consumers and construct new agents. Through empirical analysis, we seek the agents who really determine the price of market assets and the more reasonable risk aversion coefficient. According to the family life cycle theory, the author constructs the personal agent and the family agent. This paper makes an empirical analysis of asset pricing model based on long-term consumption risk. This paper selects American resident consumption data from 1960 to 2011 and Fama French25 stock portfolio returns as samples to construct a personal agent based on age grouping criteria. People and family agents based on average age, The empirical analysis of CCAPM. Taking the consumption data of personal agent as the sample, the author finds that the consumption data of individual agent over 45 years old can better explain the model. And the calculated risk aversion coefficient is more in line with the actual risk aversion coefficient. Taking the consumption data of the household agent as the sample, the author finds that the determinant of the stock market price is the family with the older average age. The calculated risk aversion coefficient is smaller than the calculated result based on the sample of total consumption risk, and the fitting degree of the model is much higher than that of other family agents. The author constructs and forecasts the consumption data of the CGF complement Growth Factor-mimicking portfolio, and makes an empirical analysis. The empirical results further confirm the above conclusions. The results of this paper show that the asset pricing model can be better explained by taking household consumption as the research object than by the individual. The real determinants of asset prices are families that do not bear the burden of raising and educating their children. Again, this paper verifies that long-term consumption risks can better estimate the price of assets. The accuracy of the asset pricing model based on long-term consumption risk is verified.
【学位授予单位】：厦门大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F126

【共引文献】