基于Black-Litterman模型的QDⅡ基金全球资产组合优化配置研究

发布时间：2018-01-02 01:00

本文关键词：基于Black-Litterman模型的QDⅡ基金全球资产组合优化配置研究　出处：《哈尔滨工业大学》2014年硕士论文　论文类型：学位论文

【摘要】：合格境内机构投资者制度(QDII)是指有限制的放开符合一定条件的大陆境内机构投资者投资境外资本市场,是在资本市场尚未完全对外开放下的临时性、过渡性的投资制度。我国自2007年开始实施合格境内机构投资者制度,这一方面为投资者提供了更加广阔的平台和投资渠道,另一方面也使投资者面临更多的风险暴露,出现了诸多亟待解决的战略战术问题。由于人才缺乏、资产配置失衡以及金融危机等国际资本市场波动的诸多方面的原因,QDII基金无论在发行份额方面还是整体收益方面均业绩惨淡,其中,资产配置过于集中、过分依赖中概股是很重要的一部分原因。本文以QDII基金的资产配置为视角,在回顾和分析了国内外关于资产配置理论的文献后,选取了主动型管理模型——Black-Litterman模型,综合市场隐含收益率与投资者主观看法,对当前的QDII基金资产区域配置进行了重新优化。在资产配置模型的运用方面,本文的主要创新在于运用多元向量自回归模型计算BL模型中的投资者观点,而并非以往研究中根据投资者主观判断随意赋值的方法。为了优化基金资产配置,增加基金的风险收益率,本文着重以下两个方面的研究与应用。首先,选取了2013年、2014年QDII基金年报中所覆盖的十个全球主要资本市场,根据已有的理论文献支持,选取了指数收益率、国民生产总值、狭义货币供应量、通货膨胀率和市场利率五个指标,利用多元向量自回归模型(VAR)对2006年9月至2011年9月间的月度数据进行拟合,在拟合模型验证稳定的条件下对2011年10月至2014年3月的十大资本市场指数收益率进行预测。其次,采用Black-Litterman模型,将前一部分的预测结果作为BL模型中的投资者观点收益向量参数,结合市场隐含均衡收益率,形成后验收益,通过线性规划求解出最优配置。为了横向的比较模型的优化结果,同时与均值方差模型优化结果进行对比,对2011年四季度至2014年一季度共计十个季度的跟踪验证,配对T检验结果显示本文采用的BL模型得出的资产配置结果在以夏普比率作为绩效指标的情形下最优,而后根据实证结果提出了相关投资建议。
[Abstract]:Qualified domestic institutional investor system (QDII) refers to the limited liberalization of mainland institutional investors who meet certain conditions to invest in the overseas capital market. It is a temporary situation in which the capital market has not been completely opened to the outside world. Since 2007, China has implemented the qualified domestic institutional investor system, which provides a broader platform and investment channel for investors. On the other hand, investors are faced with more risk exposure, there are many strategic and tactical problems to be solved. The reasons for the imbalance of asset allocation and the volatility of international capital market, such as the financial crisis, are the poor performance of QDII fund both in terms of issue share and overall income. The asset allocation is too concentrated, too much reliance on shares is a very important part of the reason. From the perspective of asset allocation of QDII funds, this paper reviews and analyzes the domestic and foreign literature on asset allocation theory. The active management model, Black-Litterman model, is selected to integrate the implied return rate of the market with the subjective view of investors. This paper re-optimizes the current QDII fund asset allocation in the application of asset allocation model. The main innovation of this paper is to use the multivariate vector autoregressive model to calculate the investors' viewpoint in BL model, rather than the method of random assignment based on investors' subjective judgment in previous studies, in order to optimize the allocation of fund assets. This paper focuses on the following two aspects of the research and application. Firstly, this paper selects ten major global capital markets which are covered by the annual report of QDII fund on 2013 and 2014. According to the existing theoretical literature, the index of return, GNP, narrow money supply, inflation rate and market interest rate are selected. The multivariate vector autoregressive model was used to fit the monthly data from September 2006 to September 2011. Under the condition that the fitting model verifies the stability, this paper forecasts the returns of the top ten capital market indices from October 2011 to March 2014. Secondly, the Black-Litterman model is adopted. The former part of the forecast results as the income vector parameters of the BL model investors view, combined with the market implicit equilibrium rate of return, after the formation of the acceptance benefit. The optimal collocation is obtained by linear programming. In order to compare the optimization results of the horizontal model and the mean variance model, the optimization results are compared with those of the mean-variance model. Tracking validation of a total of 10 quarters from the fourth quarter of 2011 to the first quarter of 2014. The results of paired T test show that the asset allocation results obtained by BL model are optimal under the condition that Sharpe ratio is taken as the performance index. Then the relevant investment suggestions are put forward according to the empirical results.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.5

【相似文献】