证券投资组合选股与优化策略应用研究

发布时间：2018-08-01 08:19

【摘要】：证券投资者(含机构及众多散户)一直在寻找更好的选股策略与算法,以获得收益最大化。但由于证券市场受众多因素影响,包括公司经营状况、政策、经济环境、战争和自然灾害等,导致股票的走势难于预测,没有人能保证一定盈利。本文的研究涉及选股和优化策略两个方面：①在大盘及单支股票无法预测、股票相对来回波动前提下,研究如何利用股票的相对波动获得超过大盘收益的一种模型；②为控制风险,对选股阶段的投资组合,研究如何科学地配置投资比例,使用户在利益最大化与风险最小化之间获得一个平衡。本文主要的研究工作概述如下： 1、在对证券市场进行基本面分析和技术分析的基础上,本文基于套利这一思想出发,挖掘一种新的组合投资盈利模型：利用股票相对波动,在股票之间来回交换实现盈利。并对该模型进行了选股算法的研究,针对选股算法需要两两配对计算股票波动性、算法复杂度为o(n2)这样一个事实,提出该算法的并行与群集计算方案,达到减少计算时间的目的。通过历史数据进行模拟测试,证实了算法的可用性和有效性。 2、给出一种关于股票行业的优化配置算法。选股阶段的投资组合分属于不同的行业,需要投资者根据行业的成长性和相关经济指标等,从宏观上把握每个行业的投资资金比例。本文首先对行业进行量化分析,为行业优化配置算法提供数据支持和参考。 3、给出投资组合中投资个体的优化配置算法。该算法模型的建立依赖众多约束条件的设置,包括行业投资比例约束、beta约束、alpha约束和预期收益约束等方面,再利用二次规划求解,以此指导投资者对个股的配置比例。对于得到的配置结果,从不同维度进行了风险评估的考察,并计算出投资组合的系统风险与非系统风险提示用户,以使得用户能在收益与风险之间取得一个折中平衡。 4、设计和实现了一个原型系统。根据股票投资盈利模型和上述研究成果,本文通过计算机软件构架、算法、网络通信、数学、金融学和运筹学等相关知识,设计和实现了一个原型系统。实践和实验证明,本文的研究工作及其相关成果能够为证券投资者在组合选股和投资比例优化配置方面提供很好的参考,具有可借鉴性。实现了一个可用的原型系统,对选股算法的有效性、配置方法的实用性进行了验证。
[Abstract]:Securities investors (including institutions and many retail investors) have been looking for better stock selection strategies and algorithms to maximize returns. However, the stock market is affected by many factors, including company management, policy, economic environment, war and natural disasters, which makes the trend of stocks difficult to predict, and no one can guarantee a certain profit. The research of this paper involves two aspects: stock selection and optimization strategy. Under the premise that stock market and single stock can not be predicted and stocks fluctuate back and forth, this paper studies how to use the relative volatility of stock to obtain a model that exceeds the return of large market. 2 in order to control the risk, the paper studies how to scientifically allocate the investment proportion in the stock selection stage, so that the user can get a balance between the profit maximization and the risk minimization. The main research work of this paper is summarized as follows: 1. Based on the fundamental analysis and technical analysis of the stock market, this paper based on the arbitrage idea, A new profit-making model for portfolio investment is developed: using relative volatility of stocks to achieve profit by swapping stocks back and forth. Based on the fact that the stock selection algorithm needs pairing to calculate the stock volatility and the complexity of the algorithm is o (N2), the parallel and cluster computing scheme of the algorithm is proposed. The purpose of reducing the calculation time is achieved. The availability and validity of the algorithm are verified by the simulation test of historical data. 2. An optimal configuration algorithm for the stock industry is presented. The portfolio of stock selection stage belongs to different industries, which requires investors to grasp the proportion of investment funds in each industry macroscopically according to the growth of the industry and related economic indicators. In this paper, the quantitative analysis of the industry is carried out to provide data support and reference for the industry optimization allocation algorithm. 3. The optimal allocation algorithm for the individual in the investment portfolio is given. The establishment of the algorithm model depends on the setting of many constraint conditions, including industry investment ratio constraint, beta constraint, alpha constraint and expected income constraint, etc. The quadratic programming is used to solve the problem so as to guide investors to allocate the proportion of individual stocks. For the obtained configuration results, the risk assessment is conducted from different dimensions, and the system risk and non-system risk of the portfolio are calculated to prompt the user. A prototype system is designed and implemented to enable the user to achieve a compromise between profit and risk. According to the stock investment profit model and the above research results, this paper designs and implements a prototype system through computer software architecture, algorithm, network communication, mathematics, finance and operational research. Practice and experiment prove that the research work and its related achievements can provide a good reference for securities investors in portfolio selection and optimal allocation of investment ratio, and it can be used for reference. An available prototype system is implemented to verify the validity of the stock selection algorithm and the practicability of the configuration method.
【学位授予单位】：东华大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F830.91

【参考文献】