Predicting the Financial Performance of Banks in the Iraqi Environment Based on Credit Risk Disclosure Data under the Applications of the Fourth Industrial Revolution (ML): An Applied Study

Abstract

This research stems from a primary objective: to improve the forecasting of the financial performance of banks listed on the Iraq Stock Exchange within the Iraqi environment. This is achieved by utilizing credit risk disclosure data from the financial statements of a sample of banks (Ashur Bank and Arab Gulf Bank), within the context of the Fourth Industrial Revolution and its associated field of machine learning. The increasing need for modern technologies compatible with big data, and the current need for advanced analytical tools that transcend traditional methods, which often yield limited accuracy in financial forecasting, is particularly relevant in a banking sector grappling with challenges related to credit risk and economic instability.

Using the Iraq Stock Exchange website and relying on quarterly reports from a sample of two commercial banks listed on the exchange during the period (2016–2022), credit risk disclosure indicators were extracted, such as the non-performing loan ratio, liquidity, loan loss provisions, leverage, shareholders' equity, and return on equity. Using machine learning algorithms to compare their predictive capabilities, the random forest algorithm was chosen, and the study concluded that disclosing credit risk can be employed to improve the prediction of the financial performance of the banks in the research sample