A novel hybrid PSO-based metaheuristic for costly portfolio selection problems

Abstract

In this paper we propose a hybrid metaheuristic based on Particle Swarm Optimization, which we tailor on a portfolio selection problem. To motivate and apply our hybrid metaheuristic, we reformulate the portfolio selection problem as an unconstrained problem, by means of penalty functions in the framework of the exact penalty methods. Our metaheuristic is hybrid as it adaptively updates the penalty parameters of the unconstrained model during the optimization process. In addition, it iteratively refines its solutions to reduce possible infeasibilities. We report also a numerical case study. Our hybrid metaheuristic appears to perform better than the corresponding Particle Swarm Optimization solver with constant penalty parameters. It performs similarly to two corresponding Particle Swarm Optimization solvers with penalty parameters respectively determined by a REVAC-based tuning procedure and an irace-based one, but on average it just needs less than 4% of the computational time requested by the latter procedures.

A hybrid level-based learning swarm algorithm with mutation operator for solving large-scale cardinality-constrained portfolio optimization problems

“…We first point out the strengths and weaknesses of using the proposed algorithm to tackle large-scale cardinality-constrained portfolio optimization problems. The comparisons are made with two recent variants of the LLSO as well as with other state-of-the-art swarm optimization algorithms implementing the exact 1 -penalty function approach proposed in [15]. Finally, we assess the profitability of the investment strategy in a real-world case study by varying the size of portfolios.…”

“…In the previous subsection, we have analysed the impact of the mutation on the capabilities of LLSObased algorithms, finding that the ALLSO-MUT-H is the more efficient choice in terms of convergence and quality of solutions. Now, the aim is to compare the above quoted optimizer with other stateof-the-art swarm optimization algorithms, namely the PSO [15] and the Firefly algorithm (FA) [49], both endowed with an exact 1 -penalty function. In this regard, the literature presents a wide range of penalty methods to tackle constraint-handling problems [33].…”

“…In this context, swarm optimization algorithms, inspired by the self-organizing interaction among agents, have become popular in portfolio optimization theory in recent years [20]. Specifically, the particle swarm optimization (PSO) algorithm has been widely employed to solve real-world financial problems since its first proposal ( [18], [53], [45], [54] and [15]) due to its effectiveness in reaching optimal solutions. However, the aforementioned algorithm does not work efficiently when the problem size is large, leading to population stagnation and premature convergence [36].…”

Hierarchical Risk Parity and Minimum Variance Portfolio Design on NIFTY 50 Stocks

“…Hybrid models are also demonstrated that combine learning-based systems with the sentiments in the unstructured non-numeric contents on the social web [8][9]. The use of multi-objective optimization, principal component analysis, and metaheuristics have also been proposed by some researchers in portfolio design [10][11][12]. Estimating volatility in future stock prices using GARCH has also been proposed in some work [13].…”

Harris Hawks Optimization with Multi-Strategy Search and Application

“…Meta-heuristic algorithms have been widely used in all aspects of social production and life. Many related research papers are published every year in the fields of production scheduling [14,15], engineering computing [16,17], management decision-making [18,19], machine learning (ML) [20,21], system control [22], and many other disciplines.…”