基于多目标规划下的风险组合投资模型的应用研究

发布时间：2018-01-05 07:30

本文关键词：基于多目标规划下的风险组合投资模型的应用研究　出处：《武汉科技大学》2013年硕士论文　论文类型：学位论文

更多相关文章：均值 绝对偏差 VaR CVaR 多目标规划

【摘要】：论文主要研究方向是投资组合风险的度量问题，马科维兹最早提出的投资组合理论作为一个开放性问题成为近现代学者研究的主要领域。其中，方差、下半方差以及绝对偏差是对收益变量波动大小的度量；VaR、CVaR则是对资产最坏损失情况的度量。另外，下行风险度量模型包括下半方差、VaR、CVaR等等。现实中，投资决策所考虑的元素有收益大小和风险大小，收益只需要考虑收入和支出的差额，即为收益。传统意义下的投资组合模型，只是针对风险度量方法的创新，而且风险度量方法的选取相对单一，所以，每种投资组合模型只能针对特定投资群体来应用。本篇论文的创新之处在于，通过考虑不同风险度量方法的性质，，为了扩大投资组合模型的适用范围，我们对几种传统风向度量方法进行组合，从而提出了三种创新模型，即均值-方差-CVaR模型、均值-绝对偏差-VaR模型、均值-绝对偏差-CVaR模型。它们所具有的优势是同时考虑了收益的波动风险和投资者最关心的资产最坏损失风险，这就意味着模型更加贴近现实情况，风险度量更加科学。尤其绝对偏差与CVaR的风险组合不仅简化了计算量，同时避免了传统VaR所不具有的次可加性，从而使模型更加有效。论文建立的几组创新模型为多目标决策问题，根据模型特点和实际需要，论文选用了多目标决策方法——约束法，投资者可根据个人偏好选择目标函数，从而增加了模型的应用范围，最后论文结合我国股票市场进行实证研究，最后得出模型可以同时降低波动风险和最坏损失风险，从而降低整体风险，并证明了模型的有效性。
[Abstract]:The main research direction is the measure of portfolio risk . The first investment portfolio theory proposed by Marcowitz is the main field of the study of modern scholars . The variance , the lower half variance and the absolute deviation are the measure of the volatility of the income variable . In the reality , the element of the investment decision includes the lower half variance , VaR , CVaR and so on . In reality , the investment portfolio model under the traditional sense only needs to take into account the difference between income and expenditure , which is the income . The choice of the risk measurement method is relatively single , so that each investment portfolio model can only be applied to the specific investment group . The innovation of this paper is that by considering the nature of different risk measurement methods , we combine several traditional wind direction measures to enlarge the application scope of the portfolio model . The advantages are : mean - variance - CVaR model , mean - absolute deviation - VaR model , mean - absolute deviation - CVaR model . The advantages are that the model is closer to the reality , and the risk measure is more scientific . Especially the combination of absolute deviation and CVaR can not only simplify the calculation amount , but also avoid the subadditivity of traditional VaR , so that the model is more effective . Based on the characteristics of the model and the actual needs , the paper selects the multi - objective decision - making method _ constraint method , and the investor can select the objective function according to the personal preference , thus increasing the application scope of the model , and finally , combining with the empirical research on the stock market of our country , finally obtaining the model can reduce the fluctuation risk and the worst loss risk at the same time , thereby reducing the overall risk , and proving the validity of the model .

【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F830.59;F224

【参考文献】