止损策略对双随机安全第一投资组合模型的影响研究

发布时间：2018-01-03 02:44

本文关键词：止损策略对双随机安全第一投资组合模型的影响研究　出处：《重庆大学》2014年博士论文　论文类型：学位论文

【摘要】：本文的主要研究内容是证券投资组合理论，针对证券市场会重复出现某些技术形态以及投资者会采用止损策略这两个现象，从如何确定股票收益率，如何建立合适的投资组合模型，如何设计高效的求解算法，止损策略如何对投资组合模型产生影响，如何确定止损策略的最优止损点和止盈点这五个方面入手，运用理论分析和定量分析相结合的方法做出了深入研究。主要研究内容可概述如下： ①大量的理论和实践表明，股票市场会重复出现某些技术形态，具有某些形态的股票容易上涨，具有某些形态的股票容易下跌。如果投资者能够总结这些重复出现技术形态的规律并将其用于选择股票，则可以提升其在股市的盈利表现。针对这一现象，，本文首先给出了一种新的量化股票收益率的方法并用双随机变量来描述股票收益率；随后，在已有文献的基础上建立了双随机安全第一投资组合模型并设计了一种融合了双重随机模拟技术与遗传算法的混合智能算法；最后，借助两类数值例子验证了新模型和算法的有效性，并根据数值结果给投资者提出相应的投资建议。 ②为了更好地求解双随机安全第一投资组合模型，本论文还设计了一种新的融合了LGMS-FOA算法和双重随机模拟技术的混合智能算法。首先，本文检验了果蝇算法（FOA）求解复杂优化问题的能力；随后，在FOA算法的基础上提出了LGMS-FOA算法；最后，将LGMS-FOA算法与双重随机模拟技术相结合得到了新算法，并将其与已有的算法进行比较。 ③人们的投资过程包括两部分，一是如何买入，二是如何卖出。然而研究卖出策略对投资组合模型影响的文献很少。止损策略是一种非常有效的卖出策略，本论文首先研究了止损策略对双随机安全第一投资组合模型的影响；随后，建立了一个带有止损策略的双随机安全第一投资组合模型并设计了一种融合了LGMS-FOA算法与双重随机模拟技术的混合智能算法；最后，给出了一个数值例子以验证模型和算法的有效性。基于以上研究，本文的结论如下： ①本论文用双随机变量来量化股票收益率，该变量可以很好地体现技术形态与投资者异质性的特点；本论文建立的双随机安全第一投资组合模型既可以兼顾风险和收益，又可以适用于所有投资者，且数值例子也证明了本文模型和算法的有效性。 ②本文通过实验发现：已有的FOA算法不能很好地求解复杂优化问题，原因是FOA算法存在一种非线性的候选解产生机制，正是这种机制限制了FOA算法求解复杂优化问题的性能。为了克服FOA算法的缺陷，本文提出了LGMS-FOA算法，并从理论和实例两个方面证明了LGMS-FOA算法要优于FOA算法，同时，在求解双随机安全第一投资组合模型时，融合了LGMS-FOA和双重随机模拟技术的智能算法二也优于融合了遗传算法和双重随机模拟技术的智能算法一。 ③本文发现：止损策略会改变投资组合的比例，当设置止损点和止盈点后，投资者需要根据止损点和止盈点来改变相应的股票收益率，否则会造成投资组合模型失效。 ④在给定股票收益率以及风险和收益的条件下，选取合适的止损点和止盈点，止损策略的表现要优于非止损策略。这是因为采取止损策略的投资者关心的是股价超过止损点和止盈点的累计概率，而不再是更高的收益率，因而会采用新的资产组合比例。
[Abstract]:The main content of this paper is to study the securities portfolio theory, the stock market will be repeated in some form of technology and investors will use the two exit from the phenomenon, how to determine the stock returns, how to establish a suitable model for portfolio investment, how to design an effective algorithm, exit strategy how to affect the portfolio model, how to to determine which of the five aspects of the optimal exit stops and profit point, using the method of theoretical analysis and quantitative analysis combination made in-depth research. The main research contents can be summarized as follows:
Show that a lot of theory and practice, the stock market will be repeated in some form of technology, with some form of stock up easily, with some form of stock to drop. If investors can summarize these repeated form of technical rules and used to select stocks, can enhance the performance of the stock market profit for this. The phenomenon, this paper proposes a new method to quantify the rate of stock return and double random variables to describe the stock returns; then, on the basis of the existing literature established random safety first portfolio model and a hybrid intelligent algorithm combines dual stochastic simulation and genetic algorithm design; finally, with the help of two kinds of numerical examples verify the validity of the new model and algorithm, and according to the numerical results for investors to put forward corresponding investment advice.
In order to better solve the double random safety first portfolio model, this paper also designs a new hybrid intelligent algorithm combines LGMS-FOA algorithm and double stochastic simulation technology. Firstly, this paper examines the Drosophila algorithm (FOA) ability to solve complex optimization problems; then, based on the FOA algorithm is put forward LGMS-FOA algorithm; finally, the LGMS-FOA algorithm and the double stochastic simulation is obtained by combining the new algorithm, and compares it with the existing algorithm.
The people of the investment process includes two parts, one is how to buy, the two is how to sell. However, research on selling strategy influences the portfolio model very little literature. Exit strategy is a very effective selling strategy, this paper studied the effect of exit strategy of double random safety first portfolio model; then, the establishment of a stop loss strategy of double random safety first portfolio model and a hybrid intelligent algorithm combines the LGMS-FOA algorithm with dual stochastic simulation design; finally, the effectiveness of a numerical example is given to verify the model and algorithm.
Based on the above research, the conclusions of this paper are as follows:
The paper uses two random variables to quantify the stock returns, the variable characteristics of technology and the heterogeneity of investors is reflected in the double random; safety first portfolio model developed in this thesis can not only take into account the risks and benefits, and can be suitable for all investors, and numerical examples demonstrate the effectiveness of this model and the algorithm.
This paper through the experiments show that FOA algorithm can not have a good reason to solve complex optimization problems, FOA algorithm has a nonlinear candidate solution mechanism, is the performance of this mechanism limits the FOA algorithm for solving complex optimization problems. In order to overcome the defects of FOA algorithm, this paper proposes the LGMS-FOA algorithm, and prove the LGMS-FOA algorithm is superior to FOA algorithm, from two aspects of theory and practice at the same time, to solve the double random safety first portfolio model, the integration of LGMS-FOA and dual stochastic simulation intelligent algorithm is better than the two intelligent algorithm combined with genetic algorithm and double stochastic simulation technique.
Third, we find that the stop loss strategy will change the proportion of portfolio. When setting stop loss point and stop point, investors need to change the corresponding stock return rate according to stop loss point and profit point, otherwise, the portfolio model will fail.
In a given stock return and risk and return, select the appropriate stops and profit, stop loss strategy's performance to be better than non stop strategy. This is because investors take exit strategy is concerned with the cumulative probability of the stock price exceeds the stops and profit point, and is no longer a higher income it will use rate, proportion of new portfolio.

【学位授予单位】：重庆大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F830.91;TP18

【参考文献】