摘要
在当前金融市场中,被动投资策略具有操作简便、成本低、风险低等一系列的优点,因此被越来越多的投资者所青睐。然而,被动投资策略的好坏主要通过跟踪误差来评判,如果跟踪误差较大,就不能够为投资者提供接近标的指数的收益。基于此本文提出了分层主成分分析方法,采用完全复制标的指数的投资策略来构建因素投资组合选择模型最小化跟踪误差,在此基础上通过改变分层的数量控制投资组合的跟踪误差。实证结果表明,基于本文提出的分层主成分分析方法,样本内与样本外的跟踪误差都很小,而且随着样本内层数的增加,样本外的跟踪误差有很大改善。
In the current financial practice,more and more investors use the passive investment strategy which has a lot of advantages,such as lower risk,lower cost and easier operation.However,the effectiveness of passive investment strategy is primarily judged by the tracking error and if the tracking error is bigger,it cannot provide the approximate return of underlying index for the investors.Thus,in this paper,the method of stratified principal component analysis which is proposed,adapts the kind of investment strategy which fully replicates the underlying index to build the factor portfolio selection model which minimizes the tracking error and controls the tracking error of the portfolio by changing the number of layer.It is showed that,based on the proposed method of stratified principal component analysis in the paper,the tracking error of between in-sample and out-of-sample is very small,and with the increase of the number of stratified layer,the tracking error of out-of-sample also will be greatly improved.
出处
《中国管理科学》
CSSCI
北大核心
2013年第S1期355-359,共5页
Chinese Journal of Management Science
关键词
被动投资策略
分层主成分分析
因素投资组合选择
跟踪误差
passive investment strategy
stratified principal component analysis
factor portfolio selection
tracking error
