CAPM的新模型研究——基于美国银行信用违约互换数据的研究

采用Jin(2011)以及Zhuo(2013)提出的新模型和新方法,对Sharpe(1964),Lintner(1965)以及Mossin(1966)论文中提出的资本资产定价模型(CAPM)进行了实证研究。本文使用极大似然法对新的模型参数进行了估计。极大似然估计的程序是用Mat Lab编写的。最后发现,对于美国银行信用违约互换而言,非对称指数幂分布模型(AEPD),以及标准化非对称指数幂分布模型(SSAEPD)都适用。通过比较发现,基于标准化非对称指数幂分布模型(SSAEPD)的资本资产定价模型具有更好的样本适用性。

Analysis of US Sector of Services with a New Fama-French 5-Factor Model

In this paper, we empirically test a new model with the data of US services sector, which is an extension of the 5-factor model in Fama and French (2015) [1]. 3 types of 5 factors (Global, North American and US) are compared. Empirical results show the Fama-French 5 factors are still alive! The new model has better in-sample fit than the 5-factor model in Fama and French (2015).

Analysis of 48 US Industry Portfolios with a New Fama-French 5-Factor Model

In this paper, we analyze US stock market with a new 5-factor model in Zhou and Li (2016) [1]. Data we use are 48 industry portfolios (Jul. 1963-Jan. 2017). Parameters are estimated by MLE. LR and KS are used for model diagnostics. Model comparison is done with AIC. The results show Fama-French 5 factors are still alive. This new model in Zhou and Li (2016) [1] fits the data better than the one in Fama and French (2015) [2].