主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2018, Vol. 26 ›› Issue (2): 107-115.doi: 10.16381/j.cnki.issn1003-207x.2018.02.012

• 论文 • 上一篇    下一篇

变动均值和方差资产配置模型的应用研究

吴文生1, 盛世杰2, 韩其恒2   

  1. 1. 合肥工业大学经济学院, 安徽 合肥 230009;
    2. 上海财经大学金融学院, 上海 200439
  • 收稿日期:2016-08-17 修回日期:2017-06-23 出版日期:2018-02-20 发布日期:2018-04-20
  • 通讯作者: 吴文生(1988-),男(汉族),安徽枞阳人,合肥工业大学经济学院,讲师,博士,研究方向:资产配置、资产定价,E-mail:ahwensheng@163.com. E-mail:ahwensheng@163.com
  • 基金资助:

    安徽省教育厅人文社会科学研究项目(JS2017AJRW0016);博士专项科研资助基金(JZ2016HGBZ1034)

Variable Mean and Variance of Asset Allocation Model in Chinese Market

WU Wen-sheng1, SHENG Shi-jie2, HAN Qi-heng2   

  1. 1. School of Economics, Hefei University of Technology, Hefei 230009, China;
    2. School of Finance, Shanghai University of Finance and Economics, Shanghai 200433, China
  • Received:2016-08-17 Revised:2017-06-23 Online:2018-02-20 Published:2018-04-20

摘要: 本文在均值-方差模型的基础上,以改善估计误差为主线,选取了10种变动均值-方差的资产配置模型,以等权重策略为基准,运用了确定性等价收益和Alpha值为判断准则,同时考虑了允许卖空限制和非允许卖空的情况,实证研究结果表明:虽然在资本市场中配置模型并不能显著战胜等权重策略,但随着投资范围的扩大,模型开始显现配置效果,尤其在Alpha准则下,变动均值-方差资产配置显著。同时本文还将实证结果和目前我国投资者的实际资产配置情况进行了比较,发现了现实配置结构中的不合理之处,并提出了相应的改善建议。最后对4类常用资产进行了模拟研究,其结果也进一步证实了本文的结论。

关键词: 资产配置, 卖空限制, 均值-方差, 估计误差

Abstract: The allocation of asset which is reasonable and effective will increase investors' wealth, and promote their consumption. This will give impetus to the development of social economy. As is known to all, the asset structure is determined by the choice of asset allocation strategy. Moreover, the constraint of borrowing and policy restriction also have significant influence on asset allocation.
Based on previous research, 10 typical asset allocation strategies under the frame of Mean-Variance Model are chosen, including equity weight strategy, improved mean-variance strategies, improved Bayesian models, and improved CAPM strategies. In this paper, these asset allocation strategies using actual data of 5 categories of most common assets in Chinese capital market are studied, which contains stocks, debts, funds, gold and real assets ranging from 2000 to 2015. All performance of these 10 strategies with and without the constraint of borrowing is also studied and compared. The empirical results with the actual asset allocations of Chinese investors are compared, and the drawbacks and unreasonable respects of the real allocations are pointed out. The result provides some corresponding improvement suggestions.
The empirical results of this paper show that since the estimation error is the largest in the capital market, the performance of theoretical optimal asset allocation is not as good as the performance of equity weight strategy. In the meanwhile, the optimal asset allocation is dominant to the equity weight strategy in the case of financial market because the estimation error is the smallest. Furthermore, the constraint of borrowing will restrain the effect of asset allocation strategies, especially in the case that gold and real assets are all included in our investment scope. First, the current investment scope of Chinese investors is too small. Investment diversification can not only increase the effect of asset allocation, but also promote the development of the market. Second, the constraint of borrowing indeed restrain the abnormal fluctuation of the market in the short term, and it will affect the asset allocation and the development of the market negatively in the long term. So the government should relax restriction, introduce more investment products and increase market depth and elasticity to attract and guide more investors into the market.
In the last part of this paper, the performance of the strategies of mean-variance and equity weight is analyzed by using the data of 4 kinds of common assets and employing the method of Monte Carlo Simulation. The result makes a further support to our previous conclusion.

Key words: asset allocation, short sellingl, mean-variance, estimation error

中图分类号: