Strong corporate governance and one-time government will reduce profit distortion, increase transparency and trust of investors and shareholders. On the other hand, the influence of political interference and weak corporate governance leads to profit distortion and correct and reliable information is not provided to investors and shareholders. Therefore, the purpose of this research is to investigate the effect of political relations on corporate governance and management accounting in the economy of Iraq with the approach of generalized moments and machine learning. This research has analyzed data related to Iraqi companies in the period from 2015 to 2023. Machine learning models including decision tree, reinforcement gradient and SVM algorithm have been used for data analysis. The results show that political relations, especially internal relations (IO), external relations (GO), and their combination (IO*GO) have a significant effect on improving the indicators of corporate governance and management accounting. The analysis of different models showed that variables related to political relations are of high importance and can help improve management decisions and the overall performance of companies. Also, the percentage of government ownership and the role of institutional shareholders play an important role in strengthening managerial accounting and corporate governance. Research findings based on advanced statistical methods and machine learning algorithms confirm that companies should pay special attention to strengthening and expanding political relationships in order to improve their performance. The results and suggestions of this research can help decision makers and company managers in Iraq in improving strategic planning and upgrading management and accounting systems.