Studies on the mathematical model and algorithm of stock-bond portfolio problem

In order to propose a optimal strategy for stock-bond portfolio problem, in this paper, the authors not only define the semi-absolute deviation risk function of stock-bond but also establish a mathematical model, which takes many actual factors in consideration such as the transaction costs and trading unit under no short-selling and no long-selling. Under this case, based on the original cuckoo search algorithm, an improved algorithm is worked out and is able to not only increase the searching speed of the optimal nest position but also enhance the stability of the optimal solution. In addition, this paper analyzes the complexity of this new algorithm, and finally a numerical example is given to illustrate the validity and efficiency of this algorithm as well as the established mathematical model. ©, 2015, Systems Engineering Society of China. All right reserved.