MultiObjective Dynamic Optimization of Investment Portfolio Based on Model Predictive Control

Melo, Maísa Kely de; Cardoso, Rodrigo T. N.; Jesus, Tales Argolo

doi:10.1137/20m1346420

Cited by 6 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Common risk models include mean-variance model and multiobjective evolutionary algorithm. We finally refer to multiobjective model predictive control [ 25 ] and Pareto optimization principle to simplify and derive our return-risk model. We obtain the optimal portfolio allocation ratio as the local optimal solution by some constraints obtained by the greedy algorithm.…”

Section: Model Preparationmentioning

confidence: 99%

Greedy Strategies with Multiobjective Optimization for Investment Portfolio Problem Modeling

Zhang

Zhou

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The ultimate purpose of portfolio investment is to reduce investment risk and improve total return on the premise of ensuring reasonable allocation of capital. In this paper, we build a quantitative model to advise on trading based on the price movement of Bitcoin and gold between 2016 and 2021; our goal is to maximize profit while minimizing risk. We mainly use greedy strategies with multiobjective optimization models. For the purpose of obtaining the correct price trend, some popular trend indicator strategies are referred to predict the future price trend in the medium and long term. In addition, we also consider people with different trading preferences and divided them into aggressive, advanced, balanced, and cautious and provided trading strategies for each of these four groups. This gives our model scalability. Finally, we analyze the sensitivity of the model and discuss the impact of trading commission costs on the model results. The model can be applicable to various investment situations.

show abstract

Section: Model Preparationmentioning

confidence: 99%

Greedy Strategies with Multiobjective Optimization for Investment Portfolio Problem Modeling

Zhang

Zhou

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…In recent years, multiobjective optimization algorithms (MOEAs) [1][2][3][4][5][6] have rapidly developed and been widely applied in various fields, such as investment portfolios [7][8], vehicle scheduling [9], water distribution systems [10], engineering design [11][12], and automotive engine calibration problems [13]. MOEAs aim to optimize multiple potential conflicting objectives simultaneously.…”

Section: Introductionmentioning

confidence: 99%

Many-objective Optimization Algorithm Based on the Similarity Principle and Multi-mechanism Collaborative Search

Gan,

Li,

Hao

2024

Preprint

View full text Add to dashboard Cite

In order to efficiently solve many-objective optimization problems and maintain a good distribution of solutions, an evolutionary algorithm based on the elimination of similar individuals (ESEA) is proposed in this paper. Considering that environmental selection based on Pareto- dominance relations often yields a small number of dominance-resistant individuals (DRIs), which are difficult to be eliminated naturally during the algorithm iteration process, these DRIs can significantly impact the algorithm's performance. To enhance convergence and stability, ESEA proposes a method to suppress the generation of DRIs by introducing a redundant protection mechanism. Through the elimination of distance-similar individuals in 2D or 3D objective spaces, the ESEA can effectively remove high-density individuals to maintain population diversity and good distribution. By introducing the ISDE+ indicator and eliminating angle-similar individuals in high-dimensional objective spaces, the proposed ESEA can mitigate premature convergence and promote exploration across different regions of the search space, thereby enhancing the likelihood of discovering diverse and potentially superior solutions. The proposed algorithm is compared with five state-of-the-art optimizers on 21 test problems. The experimental results demonstrate the promising performance of the proposed algorithm while effectively searching for the ideal Pareto front. It is evident that eliminating similar individuals and performing multiple mechanism joint searches can enhance the selection pressure towards the ideal Pareto front. Furthermore, the proposed ESEA can not only find a set of well-distributed points on the entire Pareto-optimal front, but also effectively maintain boundary solutions very well.

show abstract

Multi-indicator collaborative evolutionary algorithm for many-objective optimization

Gan,

Li,

Hao

et al. 2024

Cluster Comput

View full text Add to dashboard Cite

MultiObjective Dynamic Optimization of Investment Portfolio Based on Model Predictive Control

Cited by 6 publications

References 16 publications

Greedy Strategies with Multiobjective Optimization for Investment Portfolio Problem Modeling

Greedy Strategies with Multiobjective Optimization for Investment Portfolio Problem Modeling

Many-objective Optimization Algorithm Based on the Similarity Principle and Multi-mechanism Collaborative Search

Multi-indicator collaborative evolutionary algorithm for many-objective optimization

Contact Info

Product

Resources

About