DETERMINING THE OPTIMAL NUMBER OF BOARD MEMBERS: IMPLEMENTATION OF ARTIFICIAL NEURAL NETWORKS

Authors : Gökhan ÖZER, Yavuz Selim BALCIOĞLU, Abdullah Kürşat MERTER

Pages : 78-81

Doi:10.17261/Pressacademia.2023.1757

View : 28 | Download : 24

Publication Date : 2023-07-30

Article Type : Research Paper

Abstract :Purpose- The goal of this research is to delve into the complexities of board structure and composition within firms. Specifically, it aims to examine how various factors such as firm performance and firm-based play a role in determining the most appropriate number of board members. Methodology- A neural network model is created to identify the ideal number of board members based on financial performance metrics. Financial performance indicators insert ignore into journalissuearticles values(return on assets, return on equity, profits per share, and market to book value ratio); and firm-based variables compose the model\`s input layer insert ignore into journalissuearticles values(company age, company size, total sales, and leverage);. The output layer displays the ideal number of board members for each organization. The model\`s design has one or more hidden layers to represent the intricate interactions between the input variables and the desired output. Findings- As compared to the other factors, the significance of the return on assets variable as a predictor is much higher. At least one of the intervals is affected by each of the eight factors, and each of those eight variables has a statistically significant influence. Conclusion- Through a comprehensive analysis and review of existing literature, the study intends to shed light on the interplay between these factors and their impact on board effectiveness and decision-making. By exploring the relationship between firm-based factors and board composition, the research hopes to provide valuable insights and recommendations for firms looking to optimize their governance structure and improve their overall performance.
Keywords : Board size, optimal number of board members, artificial neural network, return on assets