Mahdi Khosroyani - Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
Farzaneh Heidarpoor - Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
Ahmad Yaghoob-nazhad - Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
Zahra Pourzamani - Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran.

A highly significant financial risk is liquidity risk. Liquidity risk management is a substantial part of Basel Recommendation no. three; with regard to the importance of this risk, this recommendation directs banks to develop and implement appropriate information systems for measuring, predicting, and controlling liquidity risks. Based on its structure, size, and features, each bank manages liquidity risk using different tools and methods. This study investigated the effectiveness of artificial neural networks in predicting liquidity risk in private Iranian banks. Relying on past studies and employing accounting information, this research developed a specific structure and architecture for a multilayer perceptron neural network; then, it predicted the liquidity risk of Iranian private banks from ۲۰۰۹ to ۲۰۱۹ using neural networks plus Matlab software. The research results revealed that artificial neural networks can be used to predict liquidity risk in private Iranian banks.

Keywords: Modelling, Artificial Neural Networks, Liquidity risk, Accounting Indicators, Private Iranian Banks