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 Two predominant methods for analyzing financial markets have been technical and fundamental analysis. However, the emergence of the Internet has altered the trading landscape. The availability of Internet and social media access plays a moderating role in information asymmetry, resulting in investors making informed decisions. Social media has turned into a source of information for investors. Through diverse communication channels on social media, investors articulate their perspectives on whether to buy or sell a stock. According to Surowiecki, the collective opinions gathered through social media frequently offer better predictions than individual opinions, a phenomenon referred to as the Wisdom of the Crowd. The wisdom of the crowd stands as an essential measure within social networks, with its potential to reduce errors and lessen information-gathering costs. In this study, we tried to evaluate the wisdom of the crowd's potential to improve stock price prediction accuracy. So, we developed a prediction model by Long Short-Term Memory based on the wisdom of the crowd. Users’ opinions in Persian about the Tehran Stock Exchange (TSE) stocks were collected from SAHMETO for eight months. The Support Vector Machine classified them into buy, sell, and neutral classes. During the research period, people mentioned 823 stocks, and 52 stocks with over 100 signals were chosen. The results of the study show that although the model presented has achieved an acceptable level of accuracy, correlations between the actual and predicted values exceeded 90%. The accuracy metrics of the proposed model compared to the base model were not improved.

Abstract Making decisions regarding capital structure is among the most challenging issues ahead for firms and the most critical decisions for their survival. On the other hand, several significant aspects, such as behavioral factors, have been overlooked in this field. Thus, the present study mainly seeks to identify the factors affecting capital structure in Iranian firms, emphasizing the role of behavioral factors. The present study employs mixed qualitative and quantitative research methods. From the qualitative point of view, capital market experts were inquired, and theoretical saturation was achieved using the snowball method. After the interviews, research components were extracted through coding. The opinions of a group of experts and managers of firms listed on the Tehran Stock Exchange were used in the quantitative section, and a structural equation form was used to perform confirmatory factor analysis on the research model. A total of 63 concepts in the form of six categories were identified at the first stage, which was reduced to 58 in the form of six categories and was confirmed after the concepts were sent back to the experts. The principal components included behavioral factors, macroeconomic factors, political factors, socio-cultural factors, firm features, and corporate governance. Results were validated through factor analysis in the quantitative portion of the study. The present study can be considered among the comprehensive studies at the construct level with an integrated approach to firms' capital structure. The emergence of behavioral finance resulted from understanding the importance of measuring human behavior as a factor with transcendent consequences for financial decisions. Hence, most behavioral finance studies are focused on observable behaviors. However, the item response theory presents an integrated method for disciplines that work with cognitive variables. Accepting opportunities for new knowledge is essential for firm decisions to respond to the mental views of financial managers.
The present study sought to identify the factors influencing firms' capital structure in Iran. The tool used in the present study reflected the elements making up the capital structure. In this regard, the notable point is how the classic criterion of structural capital components can explain financial managers' perception of decision-making. The research results in this area are interesting since we have confirmed a capital structure theory at the construct level. The conformity of the results and the obtained reliability levels indicate that this theory fits the given dimensions well. Moreover, relevant evidence indicates that senior financial managers adopt various states considering internal and external factors at the structural level, which can cause cognitive bias in decision-making.

Abstract The ability to predict the stock market and analyze market trends is invaluable to researchers and anyone interested in investing. However, this task is a challenging problem due to a large number of parameters and unpredictable noise that may affect the stock price. To overcome this issue, researchers have employed numerous approaches such as Moving Average (MA), Support Vector Machine (SVM), and Neural Networks. With technological advances, deep learning methods have become popular in processing time-series data. In this paper, we compare two recently introduced deep learning models, namely a Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), in forecasting daily movements of the Standard & Poor (S&P 500) index using the daily closing price of this index from 14/5/1991 to 14/5/2021. Results show that both models are effective and accurate in stock market prediction. In this case study, the mean squared error (MSE) and mean absolute error (MAE) for the GRU model are slightly lower than the LSTM model; hence, GRU outperformed the LSTM model despite its simpler structure. The results of this study are applicable in various instances where it is challenging to identify patterns among large volumes of unstructured data, such as medical data analysis, text mining, and financial time series modeling.