A new methodology for multi-period portfolio selection based on the risk measure of lower partial moments

Abstract

This study aims to optimize the multi-period investment portfolio model with lower partial moments (LPM) as a measure of risk under transaction costs constraint. A new method is used to calculate the LPM model. The meta-heuristic algorithm Non-dominated Sorting Genetic Algorithm II is used to solve the multi-period optimization problem. To show efficiency of the proposed method some quantitative performance measures such as skewness, conditional Sharpe ratio, modified Sharpe ratio and Jensen measure are used. The results show that in comparison to the regular method of computing LPM, the proposed method works better and improves the efficiency of portfolio optimization, especially in terms of the processing time.