联系我们

论文查询结果	返回搜索

论文编号：	13218
作者编号：	2120202857
上传时间：	2022/6/7 16:32:19
中文题目：	基于投资组合选择的再平衡策略构建与绩效对比分析
英文题目：	Rebalancing strategy construction and performance comparative analysis based on portfolio selection
指导老师：	齐岳
中文关键字：	投资组合再平衡；参数二次规划；交易成本；投资组合有效边界
英文关键字：	Portfolio rebalancing; Parametric quadratic programming; Transaction costs; Portfolio efficient frontier
中文摘要：	2022年2月底，中国结算数据显示，我国证券市场投资者已突破2亿。2022年1月底，中国基金业协会数据显示，境内公募基金数量达9428只。随着金融市场的发展和包括公募基金在内的投资者的增加，如何科学引导投资，减少投机行为，减少不必要的风险，维护投资者利益，促进市场健康发展成为重要问题。投资组合的再平衡有利于使实际投资更贴近既定投资策略，维持投资风险和收益之间的平衡，降低跟风投资的可能性，维护投资者利益，维护市场秩序。 近年来，我国再平衡策略的理论研究较少。国内外现有再平衡研究主要包括再平衡的动机和必要性、交易成本的考量以及再平衡的模型策略等。其中较有代表性的是DeMiguel等（2009）的研究，他们对基于不同模型的再平衡策略进行了比较，但未有详细构建精确求解有效边界的策略。此外，Mendes和Marques（2012）构建稳健投资组合并定义由基金经理操作的12种再平衡策略，但只考虑6只股票且未考虑再平衡成本。 本文则基于Qi（2020）的研究，构建了基于参数二次规划精确求解投资组合有效边界的3种再平衡策略，以及包括简单多样化在内的其他3种策略。改变股票数量和调仓周期，计算5、10、20、50、100、150、200、250只股票数量下和调仓周期分别为1、3、6、12个月时再平衡策略的绩效，分析何种再平衡策略绩效更优，并通过成对t检验与简单多样化策略进行绩效对比，判断构建的其他再平衡策略绩效是否优于简单多样化，从而为股票投资者提供不同股票数量和调仓周期下选取再平衡策略的建议。本文共涉及6种策略，8种股票数量，4种调仓周期，选取我国A股主板（包括中小板）1164只股票进行抽样以构建不同股票数量的投资组合，选取其2007年1月至2021年12月的月度数据进行计算，通过ExcelVBA编程实现计算。 本文的研究结果表明：第一，调仓至有效边界最小方差点的再平衡策略绩效在绝大多数情况下优于简单多样化，且优于其他几种策略，尤其是当投资组合规模较小时，或者再平衡调仓周期较短时。第二，当投资组合规模较大，且调仓周期较长时，调仓至有效边界非劣比率点的策略和调仓至有效边界三分之一处的策略具有较明显优势。第三，相对其他策略而言，简单多样化的再平衡策略绩效在不同股票数量下和不同投资组合间绩效差异不大，更为稳定，而其他策略则基本呈现“绩效高者更高，绩效低者更低”的特征。第四，除调仓至有效边界非劣比率点的策略外，随着调仓周期的延长，最终调仓总财富呈现下降趋势，说明再平衡操作是有必要的，有利于提升组合总财富绩效。并基于此对投资者提出建议。本文能够为股票市场的投资者尤其是个人投资者和公募基金提出再平衡新思路的理论参考，也在一定程度上说明了再平衡操作的必要性，对从投资组合选择角度提供再平衡思路、完善再平衡的实践具有一定的意义。 本文的创新之处在于：第一，目前基于参数二次规划精确求解有效边界并再平衡的策略并未有明确、详细的提出和计算，以Qi（2020）的研究为基础，创新性地基于参数二次规划精确求解有效边界提出新的再平衡策略，并考虑交易成本因素，进行较为完整的运算。第二，在我国再平衡理论研究较少的背景下，基于我国数据探究符合我国市场的再平衡策略，并比较不同股票数量和不同调仓周期下的再平衡策略绩效，找到不同股票数量和调仓周期下绩效优于简单多样化的再平衡策略。
英文摘要：	At the end of February 2022, China Settlement data showed that the number of investors in my country's securities market has exceeded 200 million. At the end of January 2022, data from the China Fund Industry Association showed that the number of domestic public funds reached 9,428. With the development of the financial market and the increase of investors including public funds, how to guide investment scientifically, reduce speculation, reduce unnecessary risks, safeguard the interests of investors, and promote the healthy development of the market has become an important issue. The rebalancing of the investment portfolio is conducive to guiding the actual investment closer to the established investment strategy, maintaining the balance between investment risks and returns, reducing the possibility of follow-up investment, safeguarding the interests of investors, and maintaining market order. In recent years, there have been few theoretical studies on rebalancing strategies in my country. The existing research on rebalancing at home and abroad mainly includes the motivation and necessity of rebalancing, the consideration of transaction costs, and the models or strategies of rebalancing. One of the most representative studies is the research of DeMiguel et al. (2009), who compared the rebalancing strategies based on different models, but did not construct a strategy for solving the efficient frontier in detail. In addition, Mendes and Marques (2012) constructed robust portfolios and defined 12 rebalancing strategies operated by fund managers, but only considered 6 stocks and did not consider rebalancing costs. Based on the research of Qi (2020), this paper constructs three rebalancing strategies based on parametric quadratic programming to accurately solve the efficient frontier of portfolios, and three other strategies including simple diversification. This paper changes the number of stocks and the rebalancing period, and calculates the performance of the rebalancing strategy when the number of stocks is 5, 10, 20, 50, 100, 150, 200, and 250 and the rebalancing period is 1, 3, 6, and 12 months. Then, which rebalancing strategy has better performance is analyzed. This paper also compares the performance of other rebalancing strategies with simple diversification through paired t-test to judge whether the performance of other rebalancing strategies constructed is better than simple diversification, so as to provide stock investors with recommendations for choosing a rebalancing strategy under different stock numbers and different rebalancing cycles. This paper involves a total of 6 strategies, 8 types of stock quantities, and 4 types of repositioning cycles. 1164 stocks on the main board of China A Shares (including the small and medium-sized board) are selected for sampling to construct investment portfolios with different stock numbers, and the period from January 2007 to 2021 is selected. Monthly data for December of the year is calculated. The results of this paper show that: First, the rebalancing strategy that repositions to the minimum variance point of the efficient frontier outperforms simple diversification in most cases and outperforms several other strategies, especially when the portfolio size is small, or when the rebalancing period is short. Second, when the portfolio size is large and the rebalancing period is long, the strategy of rebalancing to the non-inferiority ratio point of the efficient frontier and the strategy of rebalancing to one-third of the efficient frontier are better. Third, compared with other strategies, the performance of simple and diversified rebalancing strategies has little difference in performance under different stock numbers and between different investment portfolios, and is more stable, while other strategies basically show "higher performers have higher performance and higher performance." The lower is lower” characteristic. Fourth, in addition to the strategy of rebalancing to the non-inferior ratio point of the efficient frontier, with the extension of the rebalancing cycle, the total wealth of the final rebalancing shows a downward trend, indicating that rebalancing is necessary and is conducive to improving the performance of the total wealth of the portfolio. Based on this, suggestions are made to investors. This paper can provide a theoretical reference for investors in the stock market, especially individual investors and public funds, on rebalancing. It also explains the necessity of rebalancing operations to a certain extent, provides rebalancing ideas and strategies from the perspective of portfolio selection, also develops the practice of rebalancing. when the portfolio size is small or the rebalancing period is short, more consideration should be given to the rebalancing strategy of adjusting positions to the minimum variance point of the efficient frontier; when the portfolio size is large and the rebalancing period is short For a longer period of time, more consideration should be given to the strategy of adjusting the position to the non-inferior ratio point of the efficient frontier and the strategy of adjusting the position to one third of the efficient frontier; non-professional investors with weak portfolio selection ability can consider relatively more stable simple Diversification strategy; proper rebalancing is necessary in the investment process. The innovations of this paper are as follows: First, under the background that the existing research rarely uses parametric quadratic programming to solve and construct rebalancing strategies, this paper innovatively constructs a rebalancing strategy based on parametric quadratic programming to accurately solve the efficient frontier of portfolios, and considers transaction costs with more complete operation. Second, there is not much research on rebalancing theory in my country. Based on Chinese data, this paper explores the rebalancing strategy in line with the Chinese market, considers the rebalancing strategy of the portfolio under various stock numbers, and compares the performance of the rebalancing strategy under different stock numbers and different rebalancing periods. This paper finds rebalancing strategies that outperform simple diversification under different stock numbers and repositioning periods.
查看全文：	预览  下载(下载需要进行登录)