Analyzing family ownership structure and dividend policy using artificial neural network

• Published in: AIP Conference Proceedings
• Main Author: Kamaruddin S.A.; Subramaniam V.; Ghani N.A.M.; Rahim H.A.
• Format: Conference paper
• Language: English
• Published: American Institute of Physics Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85137720115&doi=10.1063%2f5.0094879&partnerID=40&md5=ef8f60316aa0caa358f44c130b685aae

In many parts of the world, many studies reveal that family firms dominate public corporations, particularly in emerging markets. In Malaysia, it is obvious that the majority large and growing businesses are mostly family owned in most emerging markets. In emerging countries, firms are highly owned by the controlling family shareholders. Given with this concentrated ownership in emerging market like Malaysia, the family shareholders can have ultimate power on firm's affair. They even can dominate the management and board members of firms. This paper aims to identify the importance of contributing factors of family ownership on dividend policy in Malaysia using both multilayer perceptron and radial basis function approaches. From the analysis done, the best artificial neural network model is multilayer perceptron (MLP) with 9-7-1 configurations. Based on the sensitivity analysis, debt ratio plays the most important predictors towards dividend yields, followed by retained earnings, return on assets (ROA), reinvested earnings, total assets, investment opportunities, board diversity ratio, independent director ratio and family ownership. Using the approach in this paper, it is expected that transparent responsibilities and rights incorporate strategies can be practiced successfully, so that no overlapping role problems happen. © 2022 Author(s).