An optimal solution for investment portfolio management with backtrack algorithms

Abstract

The aim of this research is to find optimal solutions to the investment portfolio management problem formulated in a bounded knapsack problem. An appropriate distribution of funds for each investment item will gain efficacy of investments and reduce risks. To implement the research, the authors focus on backtrack algorithms with several techniques such as constraint sorting, ratio sorting, and branch pruning. The experimental results have proven that all considered variants can find the optimal solution. However, their implementation times are different. The backtracking algorithm with branch and bound technique based on sorted ratios of benefit/cost will process the results fastest. In addition, an implication example of the best variant of the algorithm on the Vietnam stock market is provided. The results of this study have shown the potential application of the algorithm in high frequency trading and investing decision-making to optimize the benefits and reduce risks