Abstract
Stock trading highly contributes to the economic growth of the country. The stock trading objective is to earn profits with buy/sell/hold decisions on the set of stocks in the portfolio. The portfolio optimization problem is finding the decision sequence that leads to higher profit and lower risk. Portfolio optimization is challenging due to complex price history patterns and an uncertain environment. Incorrect decisions in stock trading lead to massive losses. The proposed Multi-Agent System for Portfolio Profit Optimization (MASPPO) aims to optimize trading profit and reduce risk with accurate predictions. The proposed model integrates the Fuzzy c-means with the Deep reinforcement learning model. The experimental datasets contain stock price history with 14,562 records. The MASPPO model maximizes the portfolio profit, intending to reduce the error. The proposed model, MASPPO, showed a Root mean squared error of 9.48 and a Mean absolute error of 2.63 and outpaced the recent models in the literature. The results proved that MASPPO maximizes the portfolio profit and is reliable.
Recommended Citation
Devi, Usha and R, Mohan
(2024)
"Multi-Agent System For Portfolio Profit Optimization For Future Stock Trading,"
Karbala International Journal of Modern Science: Vol. 10
:
Iss.
1
, Article 6.
Available at:
https://doi.org/10.33640/2405-609X.3337
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Included in
Biology Commons, Chemistry Commons, Computer Sciences Commons, Physics Commons
