Title :
Dynamic Portfolio Selection Under Higher Moments
Author :
Xu, Qi-fa ; Jiang, Cui-xia ; Kang, Pu
Author_Institution :
Shadong Inst. of Bus. & Technol., Yantai
Abstract :
In this paper, two defects in traditional portfolio introduced by Markowitz have been pointed out. Considering the higher moments risk and dynamic condition, the corresponding portfolio method has been inferred by Taylor series expansion of conditional expected utility function. The results show that the optimal dynamic portfolio weights satisfy a nonlinear system of equations, which has been solved based on the genetic algorithm. In the end, empirical analysis is conducted on international stock markets.
Keywords :
genetic algorithms; investment; stock markets; Taylor series expansion; conditional expected utility function; dynamic portfolio selection; international stock markets; nonlinear system; Conference management; Cybernetics; Genetic algorithms; Machine learning; Nonlinear dynamical systems; Portfolios; Statistics; Taylor series; Technology management; Utility theory; Dynamic portfolio; Genetic algorithm; Higher moments risk; IC-GARCHSK model;
Conference_Titel :
Machine Learning and Cybernetics, 2007 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-0973-0
Electronic_ISBN :
978-1-4244-0973-0
DOI :
10.1109/ICMLC.2007.4370565
