Iranian Journal of Finance

Abstract In today's complex economic landscape, accurately predicting events such as customer loan defaults presents a significant challenge for financial institutions. Traditional methods have shown limitations in accuracy, prompting the adoption of data-driven machine learning techniques for enhanced predictive capabilities. This study investigates the efficacy of novel machine-learning algorithms compared with linear models for predicting loan defaults at a major commercial bank. Data from over six thousand customer loan files spanning 2019 to 2022 were collected, cleaned, and clustered based on key loan indicators. The accuracy of predicting loan defaults was first evaluated using popular machine learning classification models, including LightGBM, XGBoost, Multilayer Perceptron, and Logistic Regression, and XGBoost performed best. After that, prediction accuracy was evaluated using various time-series machine learning algorithms, with a particular focus on a combined Gradient Boosting and Long Short-Term Memory (LSTM) approach. Results indicate that the combined algorithm outperforms traditional linear models, showing a substantial 40% improvement over the ARIMA algorithm in predicting loan default behavior. This study underscores the potential of advanced machine learning techniques to enhance predictive accuracy in the banking sector, offering valuable insights for risk assessment and financial decision-making.

Abstract Banks play a vital role in the economy by offering various financial services. One of their core functions is channelling surplus funds from units with excess resources to those with deficits, even when the latter have viable investment opportunities. This intermediary role is particularly crucial in developing economies, where capital markets are often underdeveloped and limited in scope. This study focuses on one of the most essential functions of banks—liquidity creation. Liquidity is generated when banks transform liquid liabilities into illiquid assets. While this process is fundamental to banking operations, it also introduces potential risks, especially when liquidity levels decline. In such cases, banks may become vulnerable to liquidity and credit risks. The capital adequacy ratio (CAR), disclosed in financial statements, serves as an important indicator of a bank’s resilience and its capacity to absorb losses and manage financial risks. This study investigates the relationship between liquidity creation and CAR using data from a sample of banks over the period 2011–2019, incorporating several control variables. The results support the financial fragility–crowding out hypothesis, indicating a negative relationship between liquidity creation and capital adequacy. Among the control variables, the deposit-to-asset ratio, non-interest income ratio, and bank size negatively influence CAR. In contrast, return on assets (ROA) shows a positive association, enhancing capital adequacy.

Abstract This empirical study investigates the impact of exchange rates on the financial soundness of Iranian banks from 1996 to 2023, utilizing financial soundness indicators, financial statement data, and macroeconomic variables. The selection of Iran is based on the significant role of exchange rate fluctuations in the financial soundness of its banking sector. Financial strategies and risk management adaptations are required as a result of these fluctuations, which have an impact on profitability, liquidity, and lending behavior. In contrast to the existing literature, which focuses on specific financial soundness indicators, this research establishes a composite metric informed by the International Monetary Fund's financial soundness indicators. This metric enables the analysis of exchange rate effects across a range of financial soundness levels. The study employs ARDL and quantile methodologies to investigate the differential effects on banks by their scale and the distinct impacts of official and unofficial exchange rates. The findings reveal intricate relationships, including inverse U-shaped dynamics between exchange rates and financial soundness through bank size, as well as varying effects across various quantiles of financial soundness. These insights provide crucial guidance for policymakers and financial institutions in stabilizing the banking sector in the face of economic uncertainties and underscore the necessity of customizing strategies to account for the bank size and the dynamics of exchange rates.

Abstract Sustainable Communications (SC) is a strategic approach in industries, especially in high-risk sectors, and has gained great importance today. This study explores the non-linear dynamics (U-shaped and S-shaped) between sustainable communication (SC) and financial performance (FP) in high-risk industries, including oil and gas, petrochemicals, mining, and transportation, listed on the Tehran Stock Exchange and Iran Fara Bourse over the period 2015–2024 (48 companies). Utilizing panel regression, cross-sectional regression, quantile panel regression, and Granger causality tests, the analysis integrates organizational learning, supply chain perspectives, and stakeholder theory. Findings confirm U- and S-shaped relationships, suggesting that moderate SC enhances FP, whereas excessive communication may undermine stakeholder trust, with sustained efforts yielding long-term benefits. Financial sector development amplifies the positive effects of SC, while ESG rating divergence exacerbates the adverse impacts of over-communication. Sustainability reporting requirements reinforce the benefits of balanced SC. Quantile regressions reveal heterogeneity, with stronger SC effects in high-performing firms. Granger causality tests indicate unidirectional causality from SC to FP. Industry-specific analyses highlight superior performance in petrochemicals and challenges in transportation. The study offers practical implications for optimizing SC, strengthening financial sector development, and standardizing ESG reporting. Future research should incorporate granular ESG data and dynamic modeling approaches.

Abstract This research identifies the determinants of trading risk (conditional variance) in the Iranian stock market over 15 years (2008-2023) using a novel hybrid approach combining spatial econometrics and machine learning algorithms. The main objective is to evaluate the superiority of hybrid models over traditional methods and to identify the roles of macroeconomic, geopolitical, behavioral factors, and firm characteristics in shaping systematic risk. The sample includes 172 companies listed on the Tehran Stock Exchange with 30,960 monthly observations and 33 explanatory variables. The methodology was implemented in three stages: First, GARCH and EGARCH models were employed to extract conditional variance and confirm the leverage effect. Second, the Spatial Durbin Error Model (SDEM) was used to decompose direct, spatial spillover, and total effects of variables while controlling for cross-sectional dependence (Pesaran CD statistic = 87.45***) and spatial autocorrelation (Moran's I = 0.4567***). Third, machine learning algorithms, including Linear Regression, SVM, Random Forest, XGBoost, LSTM, and Transformer, were applied independently and in combination with SDEM outputs. The results demonstrated a clear performance hierarchy: Linear Regression (R² = 0.4123, RMSE = 0.0987), SVM (R² = 0.5987), Random Forest (R² = 0.6789), XGBoost as the best standalone model (R² = 0.7456, RMSE = 0.0534), and Ensemble (R² = 0.7523). Hybrid models showed significant superiority: SDEM + XGBoost (R² = 0.7823, RMSE = 0.0471; 11.80% error reduction compared to standalone XGBoost and 52.3% improvement over Linear Regression), and SDEM + Ensemble (R² = 0.7867, RMSE = 0.0467) achieved optimal performance. Time-series cross-validation (average test RMSE = 0.0492) and the Diebold-Mariano test (DM = 3.456*** against XGBoost) confirmed statistical superiority. From a substantive perspective, the exchange rate with a total effect of 0.2443*** and SHAP contribution of 18.34% was identified as the most important systematic risk factor, followed by sanction intensity (total effect = 0.1274***, SHAP = 14.23%), Altman Z-score (SHAP = 15.67%), total stock index (total effect = -0.1801***, SHAP = 12.89%), and investor sentiment (total effect = 0.1001***, SHAP = 11.45%). The findings demonstrate that hybrid spatial econometrics and machine learning models improve prediction accuracy by 12-15% through extracting complementary information. Geopolitical and behavioral factors, in addition to traditional macroeconomic variables, are systematically important. Spatial spillovers constitute 15-25% of total effects, which are ignored in traditional models. This research shows that the frontier of financial risk modeling lies in the synergistic integration of economic theory and machine learning.