加入商品期货后投资组合优化研究

发布时间：2018-01-12 02:06

本文关键词：加入商品期货后投资组合优化研究　出处：《南京财经大学》2016年硕士论文　论文类型：学位论文

【摘要】：在国外,由于期货市场的成熟发展,国外研究者早已将商品期货作为投资工具的一种加入到投资组合中来规避风险。然而,由于我国期货市场起步的比较晚,国内研究者很少意识到商品期货具有良好的性能,比如正偏态、高流动性、低交易成本、与传统资产的低相关性等,而这些优点恰好能够很好将商品期货作为一种投资工具,以此来达到规避风险的目的。本文第二章主要回顾了前人的研究,发现前人的研究主要有几点不足:第一,均值-方差模型存在很多不足;第二,前人在投资组合中加入商品期货时从未考虑过商品可以被卖空这种情况。于是本文从前人研究中尚存在的问题出发,第三章主要介绍具有良好统计性质的风险度量方法-CVaR,包括CVaR的定义,计算及性质。第四章基于均值-方差模型的思想,构造均值-CVaR模型,以及在加入无风险资产后确定最优风险资产投资组合。第五章则在均值-CVaR模型下,首先从商品期货不允许被卖空这种情况出发,本文在由股票和债券组成的投资组合中加入某个商品期货来检测在市场上已经流通的商品期货是否能够改善投资组合;接着通过比较传统投资组合(只包含股票和债券这两类传统资产)和完全投资组合(包含传统资产和所有的单个商品期货)以及由传统资产和商品期货指数构成的投资组合的最优结果,来检测单个商品期货以及商品期货指数的分散化效果。其次,考虑商品期货允许被卖空这种情况,本文通过比较不同置信水平下的传统投资组合与完全投资组合的表现来确定商品期货的投资价值。倘若一个投资组合中资产的比重随着时间是稳定的,那么投资组合的管理就会给投资者带来较少的成本及精力。因此,本文还检测了在高、中、低风险水平下,加入商品期货后的投资组合中各资产的最优比重主要受何种因素影响。结论表明,在均值-CVaR的框架下,无论是商品期货不允许卖空还是商品期货允许卖空,商品期货确实能够给投资组合带来分散化效果。但是,风险水平越高,商品期货的风险分散能力越低。其次,在传统投资组合中加入若干个商品期货能够达到更好的风险分散效果。对于完全投资组合中资产比重受何种因素影响,结论是:在高风险水平下,资产比重只由资产的收益决定;在中等风险水平下,资产比重由资产的收益和资产间相关性程度决定;在低风险水平下,资产比重由资产间相关性以及资产本身的CVaR共同决定。
[Abstract]:In foreign countries, because of the mature development of futures market, foreign researchers have already added commodity futures as an investment tool to the portfolio to avoid risk. However, because of the late start of the futures market in China. Domestic researchers seldom realize that commodity futures have good performance, such as positive skewness, high liquidity, low transaction costs, low correlation with traditional assets and so on. These advantages can be used as an investment tool to achieve the purpose of avoiding risk. The second chapter reviews the previous studies. It is found that there are several shortcomings in previous studies: first, there are many shortcomings in the mean-variance model; Secondly, when the former people added commodity futures to the portfolio, they never considered the situation that commodities can be sold short. So this paper starts from the problems that existed in previous studies. The third chapter mainly introduces the good statistical nature of the risk measurement method-Cvar, including the definition, calculation and properties of CVaR. Chapter 4th based on the idea of mean-variance model. The mean-CVaR model is constructed and the optimal portfolio of venture assets is determined after adding riskless assets. Chapter 5th is based on the mean-CVaR model. First of all, commodity futures are not allowed to be short-sold. In this paper, a commodity futures is added to the portfolio composed of stocks and bonds to detect whether the commodity futures in circulation in the market can improve the portfolio. Then compare the traditional portfolio (which contains only two types of traditional assets, stocks and bonds) and the full portfolio (comprising traditional assets and all individual commodity futures). And the optimal result of portfolio composed of traditional asset and commodity futures index. To examine the diversification of individual commodity futures and commodity futures indices. Secondly, consider the situation in which commodity futures are allowed to be short sold. In this paper, the investment value of commodity futures is determined by comparing the performance of traditional portfolio and complete portfolio under different confidence levels. If the proportion of assets in a portfolio is stable over time. So portfolio management will bring less cost and energy to investors. Therefore, this paper also examines the high, medium and low risk levels. After adding commodity futures, the optimal proportion of each asset in the portfolio is mainly affected by what factors. The conclusion shows that under the framework of mean value-CVaR. Whether commodities futures are not allowed to sell short or commodities futures allow short selling, commodity futures can indeed bring diversification effect to the portfolio. However, the higher the risk level. The lower the risk dispersion ability of commodity futures is. Secondly, adding several commodity futures to the traditional portfolio can achieve better risk dispersion effect. The conclusion is: under the high risk level, the proportion of assets is only determined by the income of assets; Under the medium risk level, the proportion of assets is determined by the degree of correlation between the income of the assets and the assets; At low risk level, the proportion of assets is determined by the correlation of assets and the CVaR of assets themselves.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：F224;F724.5

【参考文献】