Firefly algorithm for cardinality constrained mean-variance portfolio optimization problem with entropy diversity...

Area of Science:

• Computational Finance
• Optimization Algorithms
• Swarm Intelligence

Background:

• Portfolio optimization is computationally challenging with realistic constraints.
• Nature-inspired metaheuristics are suitable but underutilized, especially swarm intelligence algorithms.
• No swarm intelligence approach exists for cardinality constrained mean-variance (CCMV) portfolio optimization with entropy constraints.

Purpose of the Study:

• To introduce a modified firefly algorithm (FA) for the CCMV portfolio problem with an entropy constraint.
• To address the exploration deficiencies of the standard FA in constrained optimization.
• To enhance portfolio selection by incorporating an entropy diversity constraint.

Main Methods:

• Modification of the standard firefly algorithm (FA) to improve exploration.
• Application of the modified FA to the cardinality constrained mean-variance (CCMV) portfolio model.
• Inclusion of an entropy diversity constraint to enhance portfolio diversification.

Main Results:

• The modified firefly algorithm (FA) demonstrated superior performance compared to existing state-of-the-art algorithms.
• The addition of the entropy diversity constraint further improved the optimization results.
• The proposed method effectively handles complex portfolio optimization problems.

Conclusions:

• The modified firefly algorithm (FA) is a promising approach for solving constrained portfolio optimization problems.
• Incorporating entropy constraints enhances the diversification and performance of investment portfolios.
• This research contributes a novel swarm intelligence application to financial optimization.