The Predictive Power of K-Nearest Neighbor (KNN): The Effect of Corporate Governance Mechanisms on Earnings Management

Abstract

This study investigates the long-run effect of corporate governance mechanisms on earnings management of listed companies in Nigeria and Ghana. The study uses Ant Colony Optimization (ACO) and K-Nearest Neighbor (KNN) in establishing a long-run effect of good corporate mechanisms in reducing earnings management practice by corporate managers. ACO selected four major corporate governance mechanisms: Board Procedure Index, Board Disclosure Index, Ownership Structure Index, and Shareholders’ Rights Index; these were the key corporate governance mechanisms that influence the reduction in earnings management activities. KNN produced a strong significant longitudinal effect of implementing good corporate governance mechanisms in decreasing the manipulating behavior of managers. Quality corporate governance mechanisms’ implementation reduces the opportunistic behavior of corporate managers in manipulating earnings. Therefore, the study alert policymakers the urgency in setting up appropriate policies to enhance the reduction in earnings management practices to provide accurate financial information for stakeholders’ financial decision-making. The use of ACO and KNN in the study is a great novelty, which presents a calibration and prediction of the impact of corporate governance mechanisms on earnings management showing the rate of reduction.