Recurrent neural network for dynamic portfolio selection

In this paper, the dynamic portfolio selection problem is considered. The Elman network is first designed to simulate the dynamic security behavior. Then, the dynamic covariance matrix is estimated by the cross-covariance matrices. Finally, the dynamic portfolio selection model is formulated. In addition, a numerical example is used to demonstrate the proposedmethod and compare with the vector autoregression (VAR)model. On the basis of the numerical example, we can conclude that the proposed method outperform to the VAR model and provide the accurate dynamic portfolio selection. 2005 Elsevier Inc. All rights reserved. 0096-3003/$ see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.amc.2005.08.031 * Corresponding author. Address: Institute of Management of Technology, National Chiao Tung University, 1001 Ta-Hsuch Road, Hsinchu 300, Taiwan, ROC. E-mail address: [email protected] (G.-H. Tzeng). 1140 C.-M. Lin et al. / Appl. Math. Comput. 175 (2006) 1139–1146