Detecting Community Structure in Financial Markets Using the Bat Optimization Algorithm

Abstract

A lucid representation of the hidden structure of real-world application has attracted complex network research communities and triggered a vast number of solutions in order to resolve complex network issues. In the same direction, initially, this paper proposes a methodology to act on the financial dataset and construct a stock correlation network of four stock indexes based on the closing stock price. The significance of this research work is to form an effective stock community based on their complex price pattern dependencies (i.e., simultaneous fluctuations in stock prices of companies in a time series data). This paper proposes a community detection approach for stock correlation complex networks using the BAT optimization algorithm aiming to achieve high modularity and better-correlated communities. Theoretical analysis and empirical modularity performance measure results have shown that the usage of BAT algorithm for community detection proves to transcend performance in comparison to standard network community detection algorithms – greedy and label propagation.

A Computational Comparison of Three Nature-Inspired, Population-Based Metaheuristic Algorithms for Modelling-to-Generate Alternatives

“…For calculation and optimization purposes, Yang (2009Yang ( , 2010) created three population-based metaheuristics: the Firefly Algorithm (FA), the Bat Algorithm (BA), and the Cuckoo Algorithm (CA). These three nature-inspired procedures have been shown to be more computationally efficient than the more commonly-used enhanced particle swarm, genetic algorithm, and simulated annealing metaheuristic procedures (Cagnina et al, 2008;Gandomi et al, 2011;Yang & Yeomans 2014) and have been applied to an extremely diverse spectrum of problem settings (Acharjee & Chaudhuri 2022;Aggrawal & Anuja 2022;Bangyal et al 2021;Bharathi 2022;Chandrasekaran & Simon 2014;Garg & Kumar 2021;Gopu & Venkataraman 2021;Pandey & Bannerjee 2021;Rahman et al 2019;Rautry et al 2019;Wang & Ji 2021).…”