基于改进遗传算法的开放式基金市场价值溢价研究  

Empirical Study of Value Premium for Open-end Fund Based on Modified Genetic Algorithm

作　　者：田新时[1] 刘磊[1] 

机构地区：[1]华中科技大学经济学院,湖北武汉430074

出　　处：《管理工程学报》2008年第2期160-163,共4页Journal of Industrial Engineering and Engineering Management

摘　　要：针对时间序列回归方法对Carhart四因素模型参数估计不够精确的缺陷,本文利用改进的遗传算法来优化四因素模型中的参数,获得了更为精确的模型。基于此模型,以2003年1月至2006年9月我国开放式基金月度数据为样本,对我国开放式基金市场的价值溢价进行了深入研究。研究结果表明我国开放式基金市场存在明显的价值溢价,并且高市值基金价值溢价比低市值基金更为显著,导致这一结果的主要原因是基金管理者的价值投资理念尚未形成。Due to the limitation of time - series regression method in parameters estimation of Carhart＇s four-factor model, a new method- modified Genetic Algorithm is adopted to optimize the parameters of four-factor model and a more precise model is obtained in this paper. Based on this model, the value premium of china open-end fund market is investigated by sampling monthly data of open-end fund for the period from January 2003 to September 2006. The empirical results show that the value premium is substaintial in china open-end fund market and the value premium of high market value fund is more significant than that of low market value fund, and the main reasons of the results are analyzed.

关 键 词：价值溢价:开放式基金 Carhart四因素模型 改进遗传算法 

分 类 号：F830.9[经济管理—金融学]