https://jfr.ut.ac.ir/ Financial Research Journal en daily 1 Mon, 22 Jun 2026 00:00:00 +0330 Mon, 22 Jun 2026 00:00:00 +0330 https://jfr.ut.ac.ir/article_107623.html ObjectiveVenture capitalists (VCs) are not long-term investors and generally exit a startup with a specific strategy, once their anticipated returns are secured. The timing and strategy of venture capital (VC) exit are critical factors that determine investors’ returns and significantly influence the success or failure of the new venture. Given the nascent nature of venture capital and the limited exit experience in Iran, the decision regarding the optimal exit time is a significant concern for Iranian VCs. Therefore, recognizing the importance of this decision, the present study aims to model the VC exit time by considering the factors that VCs use to determine their exit strategy from a startup. MethodsTo identify the factors influencing the timing and strategy of exit decisions, semi‑structured interviews were conducted with experts in the Iranian venture capital sector using convenience sampling. Subsequently, to assess the impact of these factors on the exit time of Iranian VCs, data were collected based on the identified variables. A questionnaire, developed based on existing venture capital studies, was distributed among all members of the statistical sample. The collected data were analyzed using the Competing Risks Survival Analysis method to simultaneously model the exit time and strategy. ResultsThe interviews identified 24 factors influencing the exit decision time and 7 exit strategies VCs had employed. In the subsequent modeling, all 24 factors were initially included as covariates. To eliminate factors with minimal impact, correlation and univariate analyses reduced the set to 16 factors, upon which a multivariate model was built. For multivariate modeling, Accelerated Failure Time (AFT) regression models were used. The results indicate that the AFT model based on the Weibull distribution is suitable for the survival data. The impact of each factor on the exit time was then assessed using this model. ConclusionThe results suggest that, across the exit strategies of Management Buyout (MBO), divestiture to a private investor, merger and acquisition (M&A), divestiture to stakeholders, and liquidation, the variable of Return on Investment (ROI)—a key factor highlighted in prior research—leads to a decrease in the exit hazard rate and consequently an increase in the investment duration. Furthermore, the variable “existence of a positive outlook for the startup’s future,” identified as a critical factor by Iranian VCs, reduces the exit hazard rate. This, in turn, prolongs the investment duration for strategies such as management buyout (MBO), divestiture to partners, divestiture to another VC, and divestiture to stakeholders. https://jfr.ut.ac.ir/article_107624.html Objective Trading strategies are crucial in investment companies as they guide decision-making processes and optimize returns. However, designing a profitable strategy within the complex and dynamic stock market environment poses significant challenges. The intricacies of market behavior and the multitude of influencing factors necessitate advanced modelling techniques. The growing availability of extensive data sets and increased computational power have facilitated the use of agent-based models, which have become essential tools for understanding economic and financial systems. The Tehran Stock Exchange often requires rapid adaptation due to severe volatility, regulatory changes, and sudden economic shifts. The choice to implement an ensemble strategy, consisting of deep reinforcement learning agents, arises from the unique challenges and opportunities of the Tehran Stock Exchange. Unlike traditional supervised learning models that make predictions solely based on historical data, agent-based models offer an adaptive approach that can respond to market changes in real-time. Another reason for selecting this strategy is its capacity to perform complex portfolio management operations. Combining multiple deep reinforcement learning agents, each with distinct strengths, the ensemble approach can leverage diverse strategies to optimize trades, manage risk, and enhance decision-making across different market conditions. Therefore, this research proposes an Ensemble strategy for algorithmic trading, leveraging deep reinforcement learning to optimize stock trading strategies that maximize returns while minimizing investment risk.   Methods This study implements an ensemble trading strategy by modelling the stock market and employing five distinct deep reinforcement learning algorithms. This ensemble strategy synthesizes each algorithm's strengths and best features, making it adaptable to various market conditions. To achieve this, Data from stocks listed in the price index of the top 50 companies on the Tehran Stock Exchange are utilized to train and test these algorithms. The performance of the trading agent, using different reinforcement learning algorithms, is subsequently evaluated and compared against the benchmark index and a traditional minimum-variance portfolio allocation strategy. The comparative analysis helps thoroughly assess the effectiveness of the ensemble approach in real-world trading scenarios.   Results From June 29, 2022, to January 20, 2024, the research implemented various trading models to gauge their performance. The ensemble strategy demonstrated a significant annual return of 47.13%, a cumulative return of 78.47%, and a risk-adjusted return of 1.56. These results indicate a superior performance over individual deep reinforcement learning algorithms, the benchmark price index of the 50 Tehran Stock Exchange companies, and the traditional minimum-variance portfolio allocation strategy. Among the individual algorithms, the Soft Actor-Critic (SAC) algorithm recorded the highest returns, with an annual return of 29.89% and a cumulative return of 47.89%. However, its higher annual volatility of 44.22% suggested weaker risk management. Conversely, the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm achieved a more balanced outcome with a risk-adjusted return of 0.92, highlighting its effective risk management alongside respectable returns. Therefore, the findings indicate that the ensemble strategy can effectively create a trading strategy that outperforms deep reinforcement learning algorithms, the price index of the top 50 companies on the Tehran Stock Exchange, and the minimum variance portfolio allocation strategy.   Conclusion The Ensemble strategy offers a robust and adaptive framework for dynamic stock portfolio management by combining the strengths of multiple deep reinforcement learning algorithms. It is a reliable trading strategy that enhances returns and effectively manages investment risks. Future improvements to this strategy also involve further integrating fundamental and macroeconomic indicators to refine its predictive accuracy. Additionally, incorporating legal and regulatory constraints into the stock market modeling process, as well as considering market participants beyond investors, could improve the realism and performance of the model. This holistic approach would provide a more comprehensive understanding of market dynamics, potentially leading to more stable and robust trading outcomes. https://jfr.ut.ac.ir/article_107625.html ObjectiveThe growing complexity of trading instruments, the expansion of financial market activities, and the increasing number of financial institutions have significantly increased the importance of risk identification and assessment. Since risk evaluation is considered a crucial issue in the financial systems of all countries, this study aims to examine the effects of shocks to political risk components, along with economic complexity, on the development of the Islamic financial (Sukuk) market in Iran.  MethodsThe data used in this research are based on the variables of the proposed model for the Iranian calendar years 1389–1401 (2010–2022). The data are seasonal and were obtained from the Central Asset Management Company of the Iranian Capital Market, the International Country Risk Guide (ICRG) database, and the MIT website. In this study, the Wald test was employed to examine the asymmetry of unexpected positive and negative shocks in the research variables affecting the development of the Sukuk market. In addition, the Nonlinear Autoregressive Distributed Lag (NARDL) approach was used to investigate and identify the nonlinear and asymmetric long-term relationships between political risk and its components, along with economic complexity, and the development of the Islamic financial market (Sukuk) in Iran. ResultsThe results of the model estimations in this study, using the Wald test and the Nonlinear Autoregressive Distributed Lag (NARDL) approach, indicate that positive and negative shocks in political risk and its components—including corruption, investment profile, government stability, external tensions, internal tensions, and socioeconomic conditions—have asymmetric effects on the development of the Iranian Sukuk market. Among the political risk components, the most influential positive shock in the long run is associated with internal tensions, with a coefficient of 11.87, whereas the most influential negative shock is related to external tensions, with a coefficient of -5.60. In addition, economic complexity has a positive long-run effect on the development of the Sukuk market. ConclusionDeveloped financial markets mobilize household savings, reduce transaction and control costs, facilitate risk diversification and information acquisition about investment projects, so if the stock market develops, the level of investment in the economy can increase. In this regard, Sukuk has become one of the most important and effective financial instruments in the modern world, contributing to a fundamental strategic shift in the outlook of global investors. Due to their distinctive features and compliance with Islamic principles, Sukuk instruments, which are based on the financial innovations of Muslim scholars, have gained considerable popularity in recent years. In this context, the findings of the present study indicate that political risk components have a long-term and asymmetric effect on the development of the Sukuk market in Iran. The results also indicate that positive and negative shocks to political risk, especially in components such as government instability, political violence, and corruption control, have different and significant effects on the Islamic financial market development index. From a theoretical perspective, this study provides a new framework for understanding the institutional dynamics of Islamic financial development by analyzing the nonlinear relationship between political risk and sukuk. At the strategic and policy-making level, the results of the study emphasize that creating institutional stability, improving governance, and reducing political uncertainty are among the most important prerequisites for the sustainable growth of the sukuk market in Iran. Accordingly, it is recommended that policymakers focus on strengthening transparency, reducing corruption, and creating support structures appropriate to the characteristics of Islamic finance to pave the way for the expansion and deepening of this market. https://jfr.ut.ac.ir/article_107629.html Objective This study aims to examine the impact of fiscal policies on the composition of the asset portfolio of the Central Bank of Iran (CBI). In particular, it investigates how fiscal policy instruments, including government expenditures, taxation, and transfer payments, influence the allocation of central bank assets. The study focuses on changes in the relative shares of assets such as foreign currencies, gold, government debt securities, and money demand within the CBI balance sheet. Understanding the interaction between fiscal policy decisions and the structure of central bank asset portfolios is important because central banks play a critical role in maintaining financial stability, managing foreign reserves, and supporting macroeconomic stability. Therefore, analyzing how fiscal policy adjustments affect asset allocation decisions may provide useful insights for improving macroeconomic policy coordination and central bank asset management strategies.   Methods This study uses annual macroeconomic data for Iran covering the period 1971–2023. To analyze the relationship between fiscal policies and the optimal allocation of central bank assets, an optimal control framework is developed to capture the dynamic interactions between fiscal policy variables and the composition of central bank assets. The model parameters are estimated using the Particle Swarm Optimization (PSO) algorithm, and the computational implementation is conducted in the Spyder environment. To evaluate the sensitivity of the CBI’s asset portfolio to fiscal policy changes, a sensitivity analysis is conducted for three major fiscal policy instruments, including government expenditures, taxation, and transfer payments. Different fiscal policy scenarios are simulated within the range of 0.5 to 1.5, which represent strongly contractionary to strongly expansionary fiscal policy conditions.   Results The empirical results indicate that fiscal policies may have a considerable influence on the composition of the CBI’s asset portfolio. The findings suggest that expansionary fiscal policies tend to increase the share of government debt securities in the asset portfolio while reducing the proportion of highly liquid assets. In contrast, contractionary fiscal policies appear to be associated with an increase in foreign reserve holdings and a reduction in debt securities. The sensitivity analysis further suggests that increases in government expenditures are likely to raise the share of debt securities and gold in the portfolio, possibly reflecting greater financing needs and a stronger preference for safe assets. Moreover, reductions in transfer payments and contractionary fiscal measures may decrease the share of certain foreign currencies, such as the US dollar and the Chinese yuan, in the central bank’s portfolio. The results also indicate that under expansionary fiscal conditions, the demand for safe assets such as gold and relatively stable currencies such as the euro tends to increase, which may reflect precautionary adjustments in the central bank’s asset allocation.   Conclusion Fiscal policies may play an important role in shaping the structure of central bank asset portfolios. Expansionary fiscal policies are likely to increase reliance on government debt securities, whereas contractionary policies may contribute to strengthening foreign reserve positions. Accordingly, effective coordination between fiscal and monetary policies could be important for maintaining financial stability and improving central bank asset management. Designing balanced fiscal policies, with attention to their potential effects on central bank balance sheets, may help limit excessive liquidity growth, reduce financial risks, and promote greater diversification in the central bank’s asset portfolio. https://jfr.ut.ac.ir/article_107630.html Objective This study aims to propose an innovative approach for forecasting stock closing prices using supervised deep learning techniques. The research seeks to capture temporal dependencies within stock market data and generate accurate and reliable predictions. By focusing on the Iranian stock market—a developing economy that has received limited attention in prior research—and analyzing 10 stocks over a period exceeding 10 years (2012–2023) using 21 input variables, the study investigates distinctive aspects of stock price prediction.   Methods The methodology involves integrating Long Short-Term Memory (LSTM) networks with deep learning techniques. Identifying a research gap in high-dimensional data, the study introduces the use of a Stack Supervised Autoencoder LSTM (SLSAE) to improve prediction accuracy. The study leverages LSTM neural networks to capitalize on their strong capabilities in modeling temporal dependencies. However, when dealing with high-dimensional data, relying on a single method may not be sufficient to achieve precise predictions. Therefore, we incorporate an LSTM Autoencoder (LAE) into the process. A key contribution of this paper is the use of the Supervised Autoencoder LSTM (LSAE), which significantly enhances prediction accuracy compared to previous methods. Furthermore, we utilize the Stack LSAE (SLSAE), which can identify and extract precise and valuable features from the data, ultimately leading to highly accurate predictions.   Results To conduct a comprehensive comparison, we utilize five different prediction models: SLSAE, LSAE, LSTM, Artificial Neural Network (ANN), and ARIMA. The results show that the more complex models, particularly SLSAE and LSAE, outperform simpler models like ARIMA and ANN in terms of accuracy. Across all stocks, SLSAE achieved the best results across various metrics. This model provided the lowest RMSE, MSE, and MAE values and the highest R² scores. The superior performance is attributed to the use of a multi-layered supervised architecture, which enables the detection of complex and nonlinear patterns in the time-series data. These features make the model highly effective in capturing and predicting fine details of closing stock prices.   Conclusion Predicting stock market prices is a significant challenge due to the inherent nonlinearity in the input data. When a wide range of variables are introduced as inputs, deep learning techniques demonstrate superior performance. To enhance the accuracy of stock market predictions, we utilized a deep learning architecture comprising SAE, LSAE, and a hybrid called SLSAE. The SLSAE method, however, stands out as it can directly learn deep features related to closing prices from raw input data. Unlike existing deep networks, which primarily focus on unsupervised feature learning to extract useful features from raw data, the SLSAE method can directly learn the deep features associated with closing prices. It employs a hierarchical structure, where high-level features related to closing prices are learned through building several SAE models from earlier low-level samples. Each SAE model ensures that the learned features significantly contribute to predicting price data in the output layer. Consequently, features associated with closing prices are gradually learned while irrelevant information is progressively reduced through the hierarchical stacking of SAE models. The results demonstrate that using this approach significantly improves the accuracy of stock price predictions and confirms the efficiency of the proposed method. https://jfr.ut.ac.ir/article_103114.html Objective This study aims to evaluate the efficiency of the Iranian capital market and examine the capability of hybrid machine learning models in predicting the direction of the Tehran Exchange Dividend and Price Index (TEDPIX). Additionally, the study seeks to assess the importance of influential factors affecting predictability and explainability within machine learning models.   Methods To analyze market efficiency at weak and semi-strong levels, data concerning the Tehran Exchange Divedend and Price Index from five years (2019 to 2023) is utilized. The proposed combined model includes an Extreme Gradient Boosting (XGBoost) model that is enhanced through hyperparameter optimization via a Genetic Algorithm (GA). The performance of this combined model is statistically compared against other machine learning algorithms, including XGBoost, Random Forest, Support Vector Machine, and Logistic Regression. Furthermore, to enhance the model's explainability and analyze the importance of input variables in predicting the index direction, the SHAP (Shapley Additive Explanations) method is employed.   Results The results indicate that the XGBoost-GA model outperforms other comparative models statistically, achieving an accuracy of 84% in predicting the direction of the Tehran Exchange Divedend and Price Index. A comparison of the results across different levels of market efficiency indicates that, at the semi-strong level, incorporating fundamental variables into the predictive model enhances forecasting accuracy. This finding reflects the influence of fundamental information on predicting the direction of the index and, consequently, suggests the presence of market inefficiency at this level. Additionally, at the weak-form level, the machine learning model based on technical data outperformed the random model, indicating the existence of market inefficiency at this level as well. Moreover, the model's explainability analysis, using SHAP, showed that the impact of variables on predicting the index direction varies based on the type of input data. In the purely technical model, factors related to price behavior and short-term fluctuations, such as the Relative Strength Index (RSI), trading volume, and moving average divergence and convergence, played a key role. In contrast, in the combined model that includes both fundamental and technical data, besides technical variables, factors such as real and legal liquidity influx, gold prices, and company financial indicators such as Return on Assets (RoA) and Return on Equity (RoE) significantly influenced predictions.   Conclusion The results of the study demonstrate that the Iranian capital market is inefficient at both weak and semi-strong levels, and the predictability of the Index is achievable using both technical and fundamental data. The analysis of input variable significance indicates that certain technical, fundamental, and macroeconomic indices play a more crucial role in predicting market behavior, which can contribute to more informed investment decisions and a better understanding of machine learning models' behavior in forecasting financial time series. https://jfr.ut.ac.ir/article_107553.html Objective The rising popularity of passive management in recent years is largely due to its advantages, such as lower management fees and reduced transaction costs. A key component of passive management is index tracking, which aims to replicate the performance of a specific index using a smaller set of assets. This paper introduces a novel robust linear optimization approach for tracking indices that is not only more reliable than existing models but also exhibits superior performance on out-of-sample data, effectively tracking indices over extended periods with minimal deviation and without requiring frequent rebalancing. The proposed robust model is highly adaptable, allowing decision-makers to account for a wide range of potential future scenarios, including both the most certain outcomes and the worst-case possibilities. The importance of this model lies in its ability to adapt more effectively to market fluctuations and help investors get closer to their return objectives.     Methods Given the NP-hard nature of index tracking problems with cardinality constraints, solving them in polynomial time is challenging. To address this computational burden, a novel heuristic approach was designed. This heuristic integrates the exploration capabilities of genetic algorithms with the focused search capabilities of local search techniques and is designed to solve the problem efficiently. Daily data from the Tehran Stock Exchange and OR-library were utilized to validate the proposed model and heuristic. For comparison purposes, a commercial solver was employed to solve both the proposed robust model and the conventional linear model commonly used in the literature. This approach was adopted instead of the developed algorithm because, although the proposed algorithm demonstrates superior performance compared to the commercial solver, it is inherently stochastic in nature. Therefore, to eliminate the influence of random and probabilistic factors on the results, a deterministic and precise commercial solver was utilized.   Results The results demonstrate that the proposed heuristic not only converges to optimal solutions for moderately sized problems but also produces portfolios that outperform those generated by commercial solvers in terms of both in-sample and out-of-sample data, all within a shorter time frame. Furthermore, given the structure of the Tehran Stock Exchange, the impact of including or excluding the ten largest stocks by market capitalization in the selected portfolio on the final results of the proposed tracking portfolio has been examined. This analysis indicates that the presence of these stocks in the portfolio can influence the performance of the tracking portfolio.   Conclusion This study aimed to develop an effective approach for index tracking by proposing a robust linear mathematical model alongside a novel heuristic algorithm. The performance of the proposed approach was evaluated against a benchmark linear model using five quantitative indicators: correlation, mean absolute deviation (MAD), root mean squared error (RMSE), standard deviation, and beta. The proposed model outperformed the benchmark model across all indicators in the out-of-sample period. In addition to these indicators, the objective function value was used to assess the performance of the proposed heuristic. The algorithm outperformed the CPLEX solver in 77 out of 88 comparisons, 11 indicators under four configurations for the Tehran Stock Exchange and the Hang Seng Index. Furthermore, an alternative tracking portfolio with a novel weighting system was introduced, which effectively reflects the status of the largest industries represented in the Tehran Stock Exchange. The performance of this alternative portfolio was compared with that of the proposed portfolio in terms of return, risk, and tracking error. The tracking error of the proposed portfolio was found to be nearly three times lower than that of the alternative approach.   https://jfr.ut.ac.ir/article_107631.html ObjectiveA comprehensive and integrated model of herd behavior can generalize the phenomenon of monetary illusion across convergent signaling channels associated with the movement variable. Given the two competing hypotheses of investors' cognitive and sentiment biases, the main issue is to use the inductive approach of nudge theory, which will provide a convergent model based on two expected and unexpected states in continuous probability functions of consumption and profit and loss prospects. Separating the inner and outer layers of the converging chains, empirical analyses provide further insights into how the signal chains can explain significant differences in the monetary inflation channels of the stock and foreign exchange markets, as well as a unique signal, and can be identified in bimodal convergent functions. The objective of the article is to develop and empirically test a comprehensive herd-behavior model that explains monetary illusion and market dynamics through cognitive and sentiment biases using a nudge-theory framework across stock and foreign exchange markets. MethodsTo simultaneously present data collection models, bivariate analyses, and factor analyses, the methodology is explained in a meta-composite model based on the inductive approach of the nudge theory. Using the DFT algorithm technique, regression matrices of discrete probability functions according to Hausman's theory (2005) have been developed in the framework of hypotheses. The data collection model is based on the Delphi-fuzzy method and the main theories of money illusion, which were obtained in the Tehran stock exchange market from the beginning of 2016 to the beginning of 2020. The data are related to linear and nonlinear fluctuations of the momentum index, stock market trading volume, individual trading volume, and the foreign exchange market, and have been used for selective coding of zero and one of Strauss and Corbin's (1988) theory. ResultsThe findings indicate a significant and simultaneous relationship between the main and secondary hypotheses of the monetary illusion phenomenon in expected and unexpected situations, which is due to investors' perception of inflation risk in the structural equation model of bivariate analyses. Factor analyses show the one-sided herd behavior of the two-sided signal chains of the currency markets and the unique signals of individuals who will exit the stock market in an unexpected situation. While in the expected situation of individuals, the two-sided convergences will be at the most optimal symmetrical points of the total transactions of individuals and legal entities in the outer and inner layers. ConclusionNudge theory can serve as a linking framework between herd behavior and monetary illusion. Using a deductive approach, it helps identify the structural factors that shape individuals’ risk perceptions and allows investor sentiment patterns to be expanded and generalized within the framework of Kahneman and Tversky’s (1979) prospect theory, particularly in economies experiencing chronic inflation. Through data collection, structural equation modeling, and factor analysis, this approach integrates optimal responses derived from foundational theories into a meta-synthesis model. It has been applied to distinguish convergent signal chains in the form of bimodal functions, which can identify channels through which monetary inflation signals are tracked in one-way and two-way movements of symmetric and asymmetric behavioral flows.  https://jfr.ut.ac.ir/article_106185.html Objective Excessive reliance on debt financing in a company’s capital structure increases its credit risk and, consequently, the likelihood of bankruptcy. Since shareholders are considered the residual claimants of the company, the financing method and capital structure composition can significantly influence expected returns and the pricing process of securities issued by the firm. Accordingly, this study aims to examine the role of the credit risk factor in asset pricing models and evaluate its explanatory power in explaining stock returns in the Iranian capital market.     Methods In this study, to obtain a comprehensive measure of credit risk, the Geske model—an advanced extension of the Merton model—was employed. Accordingly, the probability of default for companies’ total debt was first estimated based on the Geske model using numerical algorithm techniques. After calculating the probability of default, the credit risk factor was defined based on the difference in returns between companies with high and low probabilities of default. Next, the hedging regression method was used to examine the role of the credit risk factor in explaining stock and bond returns. In the next step, the credit risk factor was added to the asset pricing factor models, and by running time series regressions on a large set of test assets, the explanatory power of the extended models with the credit risk factor was evaluated and tested in comparison with conventional asset pricing models. Finally, to examine the robustness and stability of the results and to more accurately assess the predictability of credit risk factor loadings in explaining cross-sectional excess returns, a two-stage Fama-Macbeth test was used. In the first stage of this test, time-varying factor loadings for the credit risk factor were calculated using time series regressions on asset pricing factor models. Then, in the second stage, cross-sectional regression was performed for excess returns relative to the factor loadings estimated in the first stage. Finally, the credit risk factor price was determined as the average of the estimated coefficients from the cross-sectional regression. To achieve this goal, data from companies listed on the Tehran Stock Exchange and the Iranian OTC market between 2004 and 2023, and a diverse set of test assets, including portfolios sorted based on various company characteristics, were used.   Results The results of the spanning regression indicate that the credit risk factor contains unique and significant information that cannot be explained by other factors included in asset pricing models. Furthermore, the results of the time-series regression tests and model performance evaluation criteria demonstrate that incorporating the credit risk factor into multifactor asset pricing models enhances their explanatory power in explaining the returns of the test assets. Also, the results of the Fama-Macbeth test show that the time series average of the coefficients related to the credit risk factor is positive and significant, indicating a positive risk premium for this factor. These findings indicate that investors receive excess returns in exchange for accepting higher credit risk, and this factor is positively priced in the Iranian capital market.   Conclusion The findings of this study show that adding credit risk factor to asset pricing models significantly increases the power of these models in explaining fluctuations in returns of financial assets and stocks and also increases their forecasting accuracy. Also, the results indicate that credit risk, as a systematic and unavoidable factor that is a function of the company's economic environment, is reflected in stock returns by taking a positive risk premium and increases the expected return of stocks.   https://jfr.ut.ac.ir/article_102895.html Objective Financial literacy offers numerous benefits for individuals, businesses, society, and the planet in promoting sustainable finance. For individuals, it helps align their personal values and goals with financial decisions, diversifies their portfolios, reduces their environmental footprint, and contributes to positive social change. For businesses, it facilitates access to new funding sources, enhances reputation, improves risk management, and creates long-term value. For society, it strengthens social inclusion, reduces poverty and inequality, and supports human rights and democracy. This study aims to examine the drivers of financial sustainability within the framework of financial literacy enhancement. It argues that to make informed investment decisions about financial products with sustainable characteristics, private investors must possess a high level of financial literacy and sustainable financial literacy.   Methods This research is applied in terms of its objective and descriptive-analytical in terms of data collection. Information was gathered through both documentary-library and field methods. Financial literacy components were assessed using eleven dimensions and analyzed with MICMAC and Scenario Wizard software. The model analyzes the interrelationships between variables, enabling prioritization of influential factors and providing practical strategies to strengthen financial sustainability. Key factors and probable scenarios for financial sustainability within the framework of enhanced financial literacy were identified. The study utilized an expert-based sample of 20 participants, selected using the snowball sampling method.   Results The findings identified eleven core components that shape managers’ financial literacy, including understanding and analyzing financial statements, regulations related to social security insurance, labor laws, tax regulations, commercial law, stock market investment regulations, banking regulations, check-related laws, anti-money laundering regulations, and import-export regulations. These components play a vital role in financial decision-making, minimizing legal risks, and enhancing organizational performance. Tax regulations, as a framework for calculating, paying, and managing direct and indirect taxes, are influential in structuring financial activities and preventing issues arising from tax non-compliance. Moreover, commercial law, encompassing regulations on company formation and management, commercial contracts, and bankruptcy, serves as an essential legal tool for managing business relationships and activities. These components provide the necessary infrastructure to optimize financial and legal processes, mitigate risks, and ensure organizational compliance with legal requirements. Awareness and mastery of these topics lead to improved overall organizational performance and confidence in managerial decision-making.   Conclusion High financial literacy directly correlates with intelligent financial behaviors, such as optimal resource utilization and cost reduction. Societies that have widely implemented financial literacy programs have experienced significant improvements in welfare indices and financial sustainability. Managers must familiarize themselves with the structure and conditions for participating in international markets. Understanding the required conditions and permits, including commercial cards, customs procedures, and regulations related to smuggling, constitutes essential knowledge for managers. Additionally, knowledge of the laws and regulations governing free trade zones and special economic zones is essential for informed managerial decision-making, underscoring the importance of awareness and understanding in this area. https://jfr.ut.ac.ir/article_103240.html Banks play a critical role in supporting economies and financial markets through the allocation of resources. They act as financial intermediaries by bridging gaps between savers and borrowers and simultaneously serve as liquidity providers by managing the maturity structure of their balance sheet items. This dual functionality positions banks as pivotal institutions for ensuring the smooth functioning of economic systems. Financing the economy is among the primary socio-economic objectives of the banking sector. Achieving this goal requires banks to access a diverse range of financial resources, enabling them to support economic activities, manage credit, and service their obligations, thereby maintaining stability within the broader banking system. Bank financing involves generating financial resources to sustain banking operations and economic activities. According to the Asian Development Bank, financing encompasses various sources such as deposits, loans, payment services, and shareholder contributions. These resources empower banks to convert debts into liquid assets, effectively creating liquidity that is indispensable for economic operations. This capability underscores the importance of liquidity creation in ensuring resilience within financial systems. Additionally, the market power of banks, defined as their ability to attract customers and maintain competitive advantages, is closely linked to their financing structures. Banks leverage managerial and marketing strategies to secure market shares and expand their client bases, emphasizing the interconnectedness of financing and market dynamics. Through liquidity creation and risk transformation, banks play an irreplaceable role in stabilizing economies and fostering growth. However, previous research has predominantly focused on risk assessment, often overlooking the interaction between financing, market power, and liquidity creation. To address this gap, the present study explores the relationships among these factors, aiming to provide insights into their influence on banking operations and financial stability. To measure market power, the study utilized the Herfindahl-Hirschman Index (HHI), a well-established indicator of market concentration and competition. Bank financing was also analyzed using a modified HHI designed to capture the diversity of funding components, such as interbank borrowings, customer deposits, and issued securities. Liquidity creation was quantified using the index developed by Berger and Bouwman (2009), which categorizes balance sheet items based on their liquidity profiles. Data for this research were obtained from the financial statements of 10 publicly listed banks in Iran, covering the period from 2015 to 2022. Multivariate regression models employing pooled data techniques were applied to test the hypotheses. The findings reveal that diversified and efficient bank financing significantly enhances market power. Banks with diverse funding structures exhibit greater resilience and competitiveness, underlining the importance of effective financial management in optimizing market position. Furthermore, the study demonstrates that liquidity creation amplifies the positive relationship between bank financing and market power. As banks generate more liquidity, they strengthen their ability to attract deposits and improve their operational efficiency, creating a feedback loop that reinforces market dominance. This research underscores the need for policymakers to consider the interplay between financing structures, liquidity creation, and market power in their strategies. By fostering a deeper understanding of these dynamics, the study contributes to the development of policies aimed at enhancing banking efficiency and ensuring financial stability. https://jfr.ut.ac.ir/article_104067.html Objective: The main objective of this study is to analyze feedback behaviors in exchange-traded funds (ETFs) on the Tehran Stock Exchange, with emphasis on the impact of price deviations such as premiums and price Discount. This study examines how investors react to deviations in the price of fund units from the net asset value (NAV) and the impact of these deviations on the formation of positive and negative feedback behaviors. This study attempts to demonstrate how price deviations may affect investors’ trading decisions and cause the market to experience price fluctuations. In this regard, this study analyzes how investors react to different price conditions and return fluctuations, and examines their relationship with feedback behavior. Method: This study is of an applied type and aims to solve practical problems in the capital market that can be effective in the decision-making of investors and financial analysts. In terms of methodology, this research is descriptive-correlational and examines the relationships between the variables. The statistical population of this study includes all stock funds traded on the Tehran Stock Exchange during the period 2014-2024. After selecting 13 funds with superior performance during this period, the daily data were collected. The required information was extracted from the Iranian FIP site, which is a reliable source of Iranian capital market data. The models used in this study are based on the framework of Sentana and Wadhwani’s (1992) model, in which two types of traders, rational speculators and feedback traders, are considered. This model examines the effects of price deviations on investor behavior. In addition, the Liung-Box test (1978) was used to analyze lags in fund returns. In addition, the Chau, Deesomsak, & Lau (2011) model, which is an empirical version of the Santana and Vadwani model, was used to analyze the effect of price premiums and discounts on the feedback trading behavior in ETFs. Findings: The results show that price deviations have a significant effect on ETF trading behavior. In the case of price Premium, where the price of funds is higher than the NAV, positive feedback behavior is observed. This situation increases investors’ willingness to buy and prices. In other words, when the price of the fund exceeds the NAV, investors are more inclined to buy, and as a result, prices increase. This situation creates arbitrage opportunities and enhances the positive market fluctuations. In contrast, in the case of price Discount, where the price of the fund is lower than the NAV, negative feedback transactions are observed. In this case, investors are more inclined to sell funds and their confidence in the intrinsic value of the funds decreases. This reaction leads to a decrease in prices and an increase in negative fluctuations. Overall, the results indicate a significant effect of price Discount and Premium prices on the intensity of feedback behaviors in the market. In addition, the GARCH model shows that the volatility of fund returns significantly depends on price deviations and market conditions. Conclusion: This study shows that feedback trading in ETFs is affected by Premium and Discount. Price Premium increases investors' willingness to buy and price Discount increases their willingness to sell. These results indicate the significant role of price deviation from intrinsic value in the formation of investors' feedback behaviors. These findings suggest that investment fund managers adopt strategies to control price deviations and reduce severe market fluctuations. Additionally, the use of risk management mechanisms and increased information transparency can help improve fund performance and strengthen investor confidence. https://jfr.ut.ac.ir/article_104068.html Objective: This research aims to develop a novel model based on the Gradient Boosting algorithm and Transformer architecture for accurately predicting corporate valuation ratios at the time of Initial Public Offering (IPO) in the Iranian market. Contrary to traditional approaches focusing on first-day returns, this study evaluates the relative pricing of companies through three key ratios: Price-to-Book Value (P/B), Enterprise Value-to-Total Assets (EV/TA), and Enterprise Value-to-Sales (EV/S). Given the complexities of the Iranian market, including severe economic volatility, chronic inflation, economic sanctions, and regulatory restrictions, this research provides an efficient tool for investors, underwriters, and regulatory bodies to make more informed decisions. Methodology: This quantitative-applied study utilized data from 163 companies (42 listed on the Tehran Stock Exchange and 121 on the over-the-counter market - Farabourse) spanning the period 1392 to 1402 (approximately 2013-2023). This period encompasses diverse economic conditions, including severe sanctions, currency fluctuations, recession, and economic booms. Independent variables included firm-specific financial features (size, debt ratio, profitability, return on assets, profit margin, cash flow ratio, financial leverage, and weighted average cost of capital) and macroeconomic variables (inflation rate, main stock market index, sanctions index, exchange rate, and interest rate). The proposed architecture is a hybrid of Gradient Boosting Decision Trees, Long Short-Term Memory (LSTM) networks, and self-attention mechanisms, optimized through an Automated Machine Learning (AutoML) architecture search. Optimal parameters included a learning rate of 0.001, 150 decision trees, 4-time steps, and 3 self-attention layers. The model was compared with traditional methods (Linear Regression) and advanced methods (standard Gradient Boosting algorithms, Random Forest, and Recurrent Neural Networks) and validated using ten-fold cross-validation and bootstrapping. Findings: The results demonstrated the decisive superiority of the proposed model in predicting all three valuation ratios. For the Price-to-Book Value ratio, the developed model achieved an R2 of 0.853 and a Root Mean Square Error (RMSE) of 0.497, representing a 60% improvement in prediction accuracy compared to Linear Regression (R2 of 0.534). In predicting the Enterprise Value-to-Total Assets ratio, an R2 of 0.865 and a 69% reduction in Mean Squared Error (MSE) (from 0.152 to 0.047) were obtained. For the Enterprise Value-to-Sales ratio, an R2 of 0.846 was achieved, showing a 67% improvement in accuracy over traditional methods. Other advanced models also performed acceptably but did not reach the level of the proposed model. Cross-validation tests confirmed the model's high stability with a very low standard deviation (around 0.01), indicating its generalizability and reliability under various conditions. Variable importance analysis revealed that the inflation rate acts as the second most important factor across all models, with an average importance of 0.19, indicating the profound impact of macroeconomic conditions on corporate valuation in the Iranian market. Return on Assets (0.21) was identified as the most important financial performance indicator for the P/B ratio, Profit Margin (0.225) for the EV/S ratio, and Cash Flow to Assets Ratio (0.2) for the EV/TA ratio. Conclusion: The findings emphasize the importance of employing novel machine learning techniques in predicting IPO valuation ratios. The developed model, by reducing prediction error by 60% to 69%, provides an advanced tool for underwriters, investors, and fund managers to mitigate the risk of undervaluation or overvaluation. The identification of the pivotal role of the inflation rate highlights the high sensitivity of valuations to macroeconomic conditions in the Iranian market. This research contributes in three key areas: introducing a novel hybrid architecture, providing a comprehensive validation framework, and identifying the differential importance of variables for various valuation ratios under unstable economic conditions. https://jfr.ut.ac.ir/article_104071.html Objective: Herding behavior is a type of behavioral bias in financial markets that can lead to irrational decision-making and deviation from the intrinsic value of assets. If not managed, this behavior increases market volatility and can result in the formation of price bubbles. The bursting of these bubbles leads to significant losses for investors and financial institutions. Individual investors, due to limited time, attention, and access to uniform and limited information sources, are unable to analyze all stocks. Therefore, it seems likely that, during trading, they are drawn to stocks that attract more attention. The primary aim of this study is to investigate the effect of individual (retail) investor attention on herding behavior in the Tehran Stock Exchange. Furthermore, the study compares the effect of investor attention on herding behavior between individual and institutional investors, and also examines differences in trading and order data. Finally, factors such as the intensity of price limit hits, company size, momentum, and investor sentiment are analyzed in relation to this behavior. Method: To test the effect of individual investor attention on herding behavior, a panel data regression approach has been used. Herding behavior is calculated using the LSV and FHW models as upper and lower bounds of the true value of this behavior. Investor attention is measured through the Google Search Volume Index (ASVI) and abnormal trading volume (AVOL). In this model, to control for the effects of other influencing variables, institutional investor herding behavior, stock returns, the inverse of the P/E ratio, trading volume, standard deviation of stock returns, market capitalization, and information demand are included as control variables. To further examine this effect, companies are divided into two groups based on the median size (large and small companies) and median momentum (high and low momentum). The model is separately estimated for each group, and the results are compared. Finally, the effect of investor sentiment, measured by the Arms index, is also investigated, and the effect of attention on herding behavior is separately analyzed in both positive and negative sentiment conditions. Data was collected monthly, covering active companies in the Tehran Stock Exchange from 2009 to September 2023. Results: The findings of this study indicate that individual investors' attention plays a key role in the formation and intensification of herding behavior in the Tehran Stock Exchange. Increased attention is often driven by investors' preferences for specific stocks, leading to reduced diversity in trading decisions and greater behavioral convergence. The relationship between attention and herding behavior becomes more significant when this attention spreads widely across the market, with many investors relying on the decisions of others without thorough analysis. Research has shown that, under conditions of increased attention, individual investors exhibit stronger herding behavior due to limited information and higher susceptibility to market sentiment. This widespread attention directs investors towards similar decisions, ultimately resulting in herding behavior. Furthermore, psychological factors such as positive sentiment attract more attention to stocks, and the prevailing optimism among investors exacerbates herding behavior. Additionally, well-known and large companies, due to their reputation and media coverage, tend to attract more attention, creating a favorable environment for the formation of herding behavior. Similarly, stocks with low momentum are more likely to attract attention due to concerns about losses or the possibility of price reversals, further intensifying herding behavior. Overall, the findings emphasize the importance of individual investors' attention in shaping herding behavior and demonstrate that as individual investors' attention increases, the tendency for group decisions and convergence in the market grows stronger. Conclusion: Investors' attention, especially that of individual investors, is a key factor in the formation of herding behavior. Increased attention to specific stocks leads to reduced decision-making diversity and greater convergence among investors. Large companies, due to media coverage, and stocks with low momentum, due to concerns about losses, are more susceptible to this behavior. Market sentiment also strengthens herding behavior in positive conditions and intensifies emotional reactions in negative ones. In addition to individual investors, institutional investors' behavior also impacts this process, as it can act as a signal for retail investors. These findings align with behavioral finance theories, showing that investor decisions are shaped not only by information but also by psychological factors and the behavior of others. https://jfr.ut.ac.ir/article_104364.html Objective: This study aims to evaluate the sensitivity of machine learning models to input variables and identify the most significant factors influencing the prediction of the Tehran Stock Exchange index. Additionally, it compares the performance of different models in forecasting the stock index, focusing on the impact of input parameters, and offers strategies for optimizing input data and reducing model complexity. The key input variables considered include open, high, low prices, and trading volume. Method: In this research, due to the presence of a target variable (closing price), a supervised learning approach is employed. Furthermore, because the target variable is continuous, the problem is defined as a regression task. Predictions from four powerful machine learning algorithms—Linear Model (LM), Support Vector Regression (SVR), Artificial Neural Network (ANN), and Random Forest (RF)—were compared using performance evaluation metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and the coefficient of determination (R²). MAE represents the average magnitude of errors, while MSE indicates the differences between the predicted and actual values. Lower values for both metrics suggest greater accuracy of predictions. The R² statistic represents the percentage of variance in the data that is explained by the model, with values close to one indicating a higher level of accuracy. The dataset used in this study consists of six selected indices from various sectors, including pharmaceutical, automotive, financial, food industries, basic metals, and petroleum products, covering the period from 2020 to 2024 on a daily basis. Sensitivity analysis was conducted to evaluate the importance of input variables such as open, high, low prices, and trading volume on the predictions of each model. Findings: The sensitivity analysis revealed that Support Vector Regression (SVR) exhibits lower sensitivity to input parameters. This can be attributed to the use of nonlinear kernels in the SVR model, which allow it to model complex relationships between the variables in a transformed feature space, thus reducing the direct impact of input parameters. In contrast, models such as Random Forest, Artificial Neural Networks (ANN), and Linear Regression demonstrated relatively higher sensitivity to input variables. These models are more directly dependent on the input data and provide a better representation of the relative importance of the variables in predicting the output. ANN, in particular, showed superior performance in forecasting stock indices due to its ability to capture complex, nonlinear relationships effectively. Conclusion: This study contributes to identifying the most influential variables in stock index fluctuations and demonstrates that more complex models, such as Random Forest and ANN, can be more effective in providing accurate predictions due to their higher sensitivity to input data. These findings also highlight the value of sensitivity analysis in identifying significant variables and eliminating unnecessary features, which helps improve the speed and reduce the complexity of models. Ultimately, the results can assist investors and policymakers in making more informed decisions and refining prediction models for better capital market strategies. https://jfr.ut.ac.ir/article_105014.html Objective: The disclosure of non-financial information, such as board of directors’ reports, plays a vital role in transparency and investor decision-making. However, leverage manipulation—often motivated by the intent to conceal financial distress—can increase the complexity and opacity of such disclosures. This study aims to examine the relationship between leverage manipulation and the complexity of non-financial disclosures, while accounting for financial constraints and stock price crash risk. By focusing on publicly listed firms in the Tehran Stock Exchange, this research seeks to identify the mechanisms through which managers obscure non-financial information and explores the impact of such complexity on market perception. Methods: This applied study employed a descriptive-correlational approach. The statistical population comprised board activity reports from 129 companies across 19 industries listed on the Tehran Stock Exchange, covering the period from 2007 to 2023 (1386–1402 in the Persian calendar), yielding 2,193 firm-year observations. Data were extracted from the Codal database and analyzed using a panel data approach. Leverage manipulation was measured based on the Zhu (2006) model, while complexity indicators were selected through theoretical literature and factor analysis. Partial Least Squares (PLS) modeling and ordinary least squares (OLS) regression were used to evaluate the relationships, allowing for the examination of cross-sectional dependencies and complex interactions. Results: The main findings of the research proved there is a positive and significant relationship between leverage manipulation and difficulty and complexity of text of non-financial information. For the purpose of covering manipulations, some managers with various motivations make the text of non-financial information difficult, in this case leverage ratios will be less than the truth. The study revealed, there is a significant difference among industries. when it comes to leverage manipulation that some companies are likely to manipulate leverage. In terms of motivation, financial constraint and risk of stock price crash play key role in this case. Additionally, control variables have impact on modeling result which refers some factors have influence and connections with difficulty and leverage manipulation variables. Based on structural equation modeling and variables, three hypotheses are discussed in this article, all of which have been confirmed. Conclusion: This study demonstrates that leverage manipulation affects not only quantitative metrics but also the qualitative dimensions of corporate reporting. Managers may employ linguistic techniques such as passive constructions, excessive verbosity, and technical jargon to diminish the clarity of non-financial information. This tendency intensifies under conditions of financial constraints and stock price crash risk, ultimately distorting investor perception and undermining transparency. Unlike quantitative manipulations, which are often detectable via standard audit procedures, qualitative obfuscation is subtler and harder to identify—making it a critical area for scrutiny. Aligned with signaling theory, the findings show that increasing disclosure complexity functions as a negative signal, reducing investor trust and elevating information asymmetry. This research contributes to the body of knowledge by integrating financial manipulation with linguistic opacity, offering a multi-dimensional framework for understanding disclosure practices. From a practical standpoint, enhancing the financial literacy and accounting knowledge of investors is essential, as non-financial reports assume a foundational understanding of accounting concepts. In addition, policymakers are encouraged to establish clearer reporting standards and enforce disclosure simplicity to prevent undue complexity. The study’s limitations include its focus on Iranian publicly listed firms and the exclusion of international comparisons. Future research may extend this model using structural equation modeling or machine learning techniques across diverse markets and sectors. Ultimately, this research offers a theoretical and empirical framework for enhancing reporting transparency and mitigating information asymmetry in capital markets. https://jfr.ut.ac.ir/article_105890.html Objective Climate change poses an unprecedented challenge to the governance of global socioeconomic and financial systems. Our current production and consumption patterns cause unsustainable emissions of greenhouse gases (GHGs), especially carbon dioxide (CO2): their accumulated concentration in the atmosphere above critical thresholds is increasingly recognised as being beyond our ecosystem’s absorptive and recycling capabilities. The risks caused by climate change are considered to be the most serious climate risks in the next 30 years. These risks directly impact financial stability through channels such as the destruction of collateral, assets, and physical infrastructure of banks and customers caused by extreme climate events. Increasing of banks risks due to climate change can cause financial instability that pose serioes damage to the whole economy. Therefore, it has been the main focus of researchers and policy makers in various countries. In this research, banks risk due to climate change was investigated. Methods In the present study, using data from 17 Iranian banks during the period 2011-2023, using the fixed effects method and the generalized least squares (GLS) estimator, banking risk in the face of climate change was examined. Findings The results of the study show that climate change affects bank risk. The relationship between risk and precipitation is negative. This effect is direct for temperature. Regarding bank variables, the relationship between risk and capital adequacy is significant and direct. Given that the significance level of the variables of bank size, leverage ratio, deposit-to-asset ratio, and deposit-to-facility ratio is less than 0.05 percent, these variables affect bank risk. Since the coefficient of these variables is negative, this relationship is inverse. Also, the effect of bank risk on return on assets and cost-to-income ratio is not significant. Conclusion Climate change affects banks’ risk. Therefore, it is necessary for banks to strengthen their awareness of climate change risks and the physical hazards they pose. First, for industries vulnerable to extreme climate change events, such as agriculture, real estate and other industries, when banks lend to such industries, they should consider the potential risks of environmental factors and climate change to the greatest extent possible. Second, banks and related lending institutions should actively improve the disclosure of information on climate change-sensitive portfolios and enhance the monitoring of loan quality in climate-related industries to ensure that they can adapt to changing climate conditions. Finally, banks should monitor the climate characteristics of different regions, take precautions on climate change issues, and deepen their understanding of related financial laws and regulations. Coordination between the banking and insurance industries is also essential to cooperate and address the challenges of climate change. The release of accident insurance products and bank credit should be used to transfer this risk and reduce the pressure on economic agents. Similarly, increasing the penetration of insurance, reducing credit risks related to climate change, and the ability to recover quickly after disasters can be issues that need to be considered in this context. https://jfr.ut.ac.ir/article_106083.html Abstract Objective: Selecting an optimal combination of assets within an investment portfolio has always been one of the fundamental and challenging issues in investment management. In other words, investors continually aim to achieve an efficient and optimal allocation by considering the key factors influencing the decision-making process and aligning the selection with their individual risk–return preferences. However, traditional portfolio theory often relies on simplified statistical assumptions—particularly the assumption of normally distributed returns—that are inconsistent with the empirical characteristics of financial markets. In reality, financial asset returns frequently display heavy tails, negative skewness, volatility clustering, and asymmetric dependence during periods of market stress. These stylized facts imply that extreme losses occur more frequently than predicted by classical models, highlighting the necessity for more sophisticated tools in modern risk management and portfolio construction. To address these limitations, the present study develops a comprehensive hybrid modeling framework that integrates three advanced components: EGARCH to capture time-varying volatility and leverage effects, Extreme Value Theory (EVT) to model tail behavior and estimate extreme losses more accurately, and t-Copula to model tail dependence and joint downside risk across industries. Additionally, considering that parameter estimates obtained from historical data are subject to uncertainty and may deviate from their true values, a Robust Optimization approach under the Conditional Value-at-Risk (CVaR) criterion is adopted. This framework allows the portfolio to remain resilient against estimation errors and improves decision-making under market uncertainty. Accordingly, the main objective of this research is to present a robust and realistic portfolio optimization model based on the EGARCH–EVT–t-Copula–Robust CVaR approach. Methods: Daily return data for ten major industry indices of the Tehran Stock Exchange, covering the period from September 2015 to September 2025, were employed. Conditional volatility was modeled using the EGARCH(1,1) specification to account for asymmetry in the volatility response to positive and negative shocks. Standardized residuals obtained from the volatility model were analyzed using EVT under the Peaks-over-Threshold method to estimate the heavy-tail behavior of returns. The dependence structure among industries was then modeled through t-Copula to capture joint extreme movements and tail dependence. Finally, portfolio optimization was conducted under the CVaR measure in both standard and robust formulations, and the performance of the resulting portfolios was compared based on expected return, risk, and Sharpe ratio. Results: Empirical findings reveal that the integrated EGARCH–EVT–t-Copula–Robust CVaR framework delivers superior accuracy in capturing extreme risks and provides a more effective portfolio allocation in heavy-tailed environments. The EGARCH model successfully captured conditional volatility dynamics, EVT highlighted the presence of heavy left-tail behavior and significant downside risk, and the t-Copula confirmed strong tail dependence among industrial sectors, implying the synchronized occurrence of severe negative shocks. In the optimization step, the Robust CVaR portfolio outperformed the standard CVaR model by yielding higher expected returns and Sharpe ratios, illustrating its enhanced ability to handle parameter uncertainty and asymmetric market volatility. Conclusion: Overall, the findings confirm that in emerging markets such as Iran—characterized by high kurtosis, negative skewness, and tail dependence—classical risk models based on normality assumptions are inadequate. In contrast, the proposed EGARCH–EVT–t-Copula–Robust CVaR framework provides a more realistic estimation of tail risk and leads to a more stable and efficient capital allocation. Hence, adopting robust CVaR optimization under conditional heteroskedasticity and tail-dependent structures is recommended for professional portfolio management and investment decision-making in high-risk markets. https://jfr.ut.ac.ir/article_106814.html Pricing Catastrophe Reinsurance Securities for Earthquake Insurance in Iran Abstract Objective: Given the importance of insurance-linked securities (ILS) as effective instruments for the development of financial markets, the main objective of this study is to price catastrophe reinsurance securities related to earthquake risks in Iran. Unlike most previous studies—especially in the Iranian context—that have primarily focused on theoretical frameworks and preliminary feasibility analyses, the present research addresses one of the most challenging aspects of the topic by emphasizing the practical pricing of these instruments. Relying on real-world data, this study examines the pricing of catastrophe bonds under various conditions, including single-period and multi-period structures. Additionally, the impact of different levels of risk premiums is analyzed, taking into account the heterogeneity in investors’ risk preferences. Method: This research is classified as applied research in terms of its objectives and as analytical research in terms of its analysis methods. The study utilizes real data from the Seismology Center of the Geophysics Institute at the University of Tehran, covering the years 2006 to 2018 (considering data accessibility limitations). Additionally, to calculate the bond prices, a standard risk-neutral pricing tool suitable for Iran's infrastructure has been employed, with the analysis conducted using R software. Findings: In this study, we aimed to calculate the prices of catastrophe bonds related to earthquakes in both a single-period (one year) and a multi-period (five years) framework using equilibrium pricing theory. The results indicate that as the frequency of repetitions within the pricing model structure increases, the price variance decreases significantly, suggesting that the model is both consistent and computationally efficient. It is worth noting that although the decline in bond prices with increasing maturity is a theoretically predictable phenomenon, the findings of this study, based on real-world data, indicate that this decline is significantly more pronounced and meaningful over longer-term horizons than what is merely suggested by theory. This insight serves as an important warning for financial instrument designers in the insurance industry and potential investors, emphasizing the need for greater caution in selecting bond maturities to avoid a substantial reduction in expected returns. Additionally, bond prices are also reduced when the risk-reward ratio component increases. In other words, the lower the risk-reward ratio, the more willing buyers are to pay a higher price for the bonds. Conclusion: The results of this research enabled the precise calculation of the prices of catastrophe bonds specifically for earthquakes. By comparing the one-period and multi-period models, as well as considering various risk premiums, the findings align with existing theories on security pricing. Given the necessity of this discussion, especially in light of Iran's earthquake risk, this model presents an effective approach for pricing catastrophe insurance securities related to earthquakes. It allows for the transfer of earthquake insurance risk from the insurance market to the capital market, capitalizing on the potential to attract risk-seeking investors who are looking for appealing investment opportunities. Overall, the issuance of insurance securities not only addresses liquidity challenges for insurance companies but can also serve as a means to reduce liquidity in society as a whole, thereby helping to control inflation. Keywords: Insurance-linked securities, pricing of securities, risk-neutral valuation, risk transfer. https://jfr.ut.ac.ir/article_106856.html The increasing volatility of financial markets and the growing interdependence among economic entities have intensified the need to pay special attention to systemic risk in investors’ decision-making processes. Systemic risk, defined as the risk of widespread disruptions that can affect the entire market or large segments of it, holds particular importance in portfolio management and financial policymaking. In this context, firm size, as a key structural factor, plays a decisive role in determining a company's vulnerability to widespread shocks. Smaller firms are generally more exposed to market fluctuations and financial crises, whereas larger firms, due to greater financial resources and operational diversification, tend to exhibit higher resilience against such shocks. A review of the literature indicates that, although various measures for assessing systemic risk have been developed in recent years, the simultaneous integration of systemic risk and firm size in portfolio optimization models has been relatively limited. Most previous studies focused primarily on individual asset risk or expected returns, often overlooking the moderating role of structural firm characteristics, particularly size, in shaping the optimal portfolio composition. This research gap motivates the design of an integrated framework that can simultaneously account for the effects of systemic risk and structural differences among firms in investment decision-making. The present study aims to provide an integrated framework for portfolio optimization in the Tehran Stock Exchange, where systemic risk is measured using the Marginal Expected Shortfall (MES) metric, firm size is considered as a moderating variable for systemic risk, and the Particle Swarm Optimization (PSO) algorithm is employed to determine the optimal portfolio composition. The use of MES as an advanced indicator allows for a quantitative assessment of the potential impacts of severe market shocks on each firm, while clarifying the role of firm size in mitigating these effects. The study’s data consist of daily stock prices and market capitalization information of listed companies over the research period. After calculating MES for each firm, companies were categorized into three groups—small, medium, and large—based on size, to examine the differential effects of systemic risk across firms with varying financial and operational structures. Subsequently, a multi-objective optimization model aimed at minimizing systemic risk and maximizing expected return was executed using the PSO algorithm, determining the optimal portfolio composition while considering real-world constraints and market conditions. The results reveal that smaller firms exhibit significantly higher systemic risk compared to larger firms, which substantially influences the structure of the optimal portfolio. In the presence of systemic risk, the weight of assets from smaller firms is reduced, and the portfolio composition shifts toward more stable, larger firms. Conversely, the share of large firms in optimal portfolios increases, indicating that investors, when accounting for systemic risk, prefer firms with greater stability and higher resilience against widespread market fluctuations. These findings emphasize that neglecting systemic risk can lead to inefficient weight allocation and inaccurate portfolio risk estimation, exposing investment decisions to considerable threats. By simultaneously integrating systemic risk, firm size, and the PSO algorithm, this study offers an innovative framework for portfolio optimization in volatile markets, serving as a practical guide for investors, asset managers, and financial policymakers. Furthermore, it lays the groundwork for future research in risk management and portfolio optimization, demonstrating that attention to both structural firm characteristics and systemic risk can enhance investment decision quality and the stability of financial markets. https://jfr.ut.ac.ir/article_107195.html Objective This study aims to forecast the performance of equity funds using supervised machine-learning algorithms. It seeks to identify the key drivers of fund performance and to propose an advanced forecasting framework that enables investors to pinpoint funds capable of generating positive alpha. In doing so, investors can make more informed capital-allocation decisions and achieve higher returns relative to passive or underperforming funds. Beyond benefiting investors, the approach can enhance overall market efficiency and the optimal allocation of capital. Methods From a research-purpose perspective, the study is both developmental and applied. We collected and cleaned data on 23 variables for 12 equity funds. Supervised learning models—linear (linear regression and elastic net) and tree-based (random forest and gradient boosting)—were implemented in Python. To maximize predictive accuracy and mitigate overfitting, hyperparameters for each algorithm were tuned via cross-validation. The dataset was split into training (80%) and test (20%) partitions. After fitting models with the optimized hyperparameters, out-of-sample performance was evaluated on the held-out test set using three accuracy metrics: mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE). For model interpretability and feature attribution, we employed Shapley Additive Explanations (SHAP). Relative predictive accuracy across algorithms was assessed using the Diebold–Mariano test. Results Tree-based models (gradient boosting and random forest) significantly outperformed linear models (linear regression and elastic net) on the evaluation metrics for 11 of the 12 funds. The value-added variable and the market return emerged as the most important return drivers across nearly all models and showed the strongest association with changes in alpha. Additionally, variables such as total net asset (TNA) and fund age exhibited pronounced importance in linear models, indicating that fund size and track record contribute to performance persistence. By contrast, nonlinear models placed greater emphasis on features like value-added and tended to down-weight many of the remaining variables. Conclusion The results underscore the strong capability of tree-based machine-learning models (gradient boosting and random forest) to analyze financial data and uncover latent patterns. Their validity, however, hinges on sound evaluation design—most notably cross-validation and careful hyperparameter tuning. As SHAP facilitated a comparative view of the determinants of fund performance in linear versus nonlinear settings, the observed superiority of gradient boosting and random forest in this study aligns with their capacity to capture nonlinearities and complex interactions. Although the magnitude and ranking of feature importance vary across models, several variables consistently play a pivotal role in explaining fund performance. A major contribution of this research is to offer methodologies that, in addition to individual investors, fund managers, financial advisors, and institutional investors—such as banks, insurance companies, and pension funds responsible for large pools of capital—to more accurately identify superior funds and optimize their portfolios using machine-learning algorithms. The proposed models may also intensify healthy competition between active and passive investment funds. https://jfr.ut.ac.ir/article_107529.html Forecasting price trends in financial markets—particularly in volatile and nonlinear environments such as cryptocurrency exchanges—has long posed a significant challenge for analysts, traders, and researchers. Among the tools commonly used for technical analysis, Japanese candlestick charts provide a powerful visual framework by depicting the behavior of prices over time. However, the interpretation of these candlestick patterns is often subjective and heuristic, lacking structured numerical criteria and reproducibility. This limitation reduces the reliability of traditional candlestick analysis. To address this issue, the present study introduces a fuzzy logic-based framework for modeling candlestick structures and predicting short-term price movements. The proposed model serves as a bridge between human intuition and algorithmic precision by leveraging fuzzy logic to quantify and formalize uncertain, linguistic concepts such as “short body” or “long upper shadow.” The study analyzes hourly Bitcoin data from September 2017 to May 2025. Each candlestick is decomposed into body, upper shadow, and lower shadow, which are standardized and fuzzified using triangular membership functions tailored to Bitcoin’s volatility. Candles are then encoded into unique three-character fuzzy codes based on dominant membership values. These codes are linked to the next candle’s trend (bullish, bearish, or neutral), enabling the extraction of statistical trend tendencies. Two predictive models are built: a soft rule model, which predicts a trend if 3 out of 4 top-matching codes agree, and a hard rule model, which requires full agreement. Both models are tested on out-of-sample data and used to simulate hourly trading strategies. Backtesting simulations are carried out under both fixed-capital and capital-adjusted strategies, assuming an initial investment of $100,000. In the soft rule model, over 10,891 trades were executed, resulting in a net profit of $190,419 with a win rate of 54.1% and an average profit per trade of $17.48. The hard rule model, with 6,761 trades, yielded a net profit of $187,979, a higher win rate of 55.8%, and a higher average profit per trade of $27.80. Both models significantly outperformed the benchmark buy-and-hold strategy. To assess the statistical significance of the observed outperformance, a paired t-test was conducted using 1,000 randomly selected 14-day intervals (336 hourly candles each). The test compared the fuzzy model returns against the buy-and-hold returns for the same periods. The resulting p-values were <0.00001 for both soft and hard rule models, confirming the superiority of fuzzy strategies with more than 99.9% confidence. The comparison between the two fuzzy models yielded a p-value of 0.463, indicating no statistically significant difference in average returns between them—though the hard rule model demonstrated a higher Sharpe ratio and lower volatility. In addition to performance evaluation, frequency analysis of fuzzy codes revealed recurring patterns associated with specific trend directions. These statistically dominant bullish and bearish fuzzy patterns could serve as a foundation for rule-based market forecasting systems. They also offer a promising opportunity for expanding the fuzzy knowledge base through semantic rule integration and asset-specific customization. Overall, this study introduces a fully interpretable, standalone candlestick-based trading model that does not rely on external indicators or multi-candle sequences. The entire system—from feature extraction and fuzzification to decision-making—is transparent and reviewable, making it suitable for financial analysts seeking clarity and trustworthiness. Compared to black-box machine learning models, the fuzzy approach offers a balanced compromise between interpretability, adaptability, and predictive power. https://jfr.ut.ac.ir/article_107661.html Objective: The creation of diversification and optimal asset allocation within a portfolio has consistently represented a complex decision-making challenge. In this study, projects are integrated as a novel asset class within the investment domain, alongside conventional risky assets such as securities and Exchange-Traded Funds (ETFs). Subsequently, utilizing a mathematical model rooted in the principles of post-modern portfolio theory, an optimal combination of all assets within the portfolio is derived. Methodology: The mathematical model employed is based on the Omega ratio. To enhance its alignment with real-world scenarios, various constraints, including boundary, cardinality, and logical constraints, are incorporated. The model is examined within a single-period probabilistic framework, utilizing the Omega ratio as the objective function. Given the nature of this ratio, its maximization is achievable solely through iterative approaches. Consequently, the problem is solved through the integration of Monte Carlo Simulation (MCS) with a metaheuristic algorithm. The solution process begins by feeding probability distribution functions—derived from expert interviews for projects and historical data for exchange-traded assets—into the MCS. This generates 5,000 scenarios, producing expected return rates for each asset. Subsequently, these return rates are combined with asset weights proposed by the metaheuristic algorithm, and the resulting Omega ratio is reported back to the algorithm. Based on the received response and its predefined structure, the algorithm modifies the asset weights to maximize the Omega ratio. This iterative process continues until the algorithm reaches a predetermined number of iterations. In this research, the Marine Predators Algorithm (MPA) is implemented as the metaheuristic algorithm. To optimize its performance, the initial parameters of the algorithm are selected using the Taguchi Design of Experiments (DOE) technique. The MPA, developed in 2020, is founded upon the theory of survival of the fittest, wherein predators with a superior ability to locate prey thrive. Results: The remarkably small error exhibited by the MPA in the benchmark test using the S&P 100 index, when compared to other metaheuristic algorithms such as Genetic Algorithm (GA), Tabu Search (TS), Simulated Annealing (SA), Particle Swarm Optimization (PSO), Population-Based Incremental Learning (PBIL), the hybrid of Population-Based Incremental Learning and Differential Evolution (PBIL-DE), and the Adaptive Ranking Multi-Objective Particle Swarm Optimization (ARMOPSO), suggests the notable validity and efficient performance of this algorithm within the portfolio optimization domain. In the numerical example presented, this algorithm effectively circumvented the penalty functions of the problem and adequately interacted with MCS, a novel asset class (projects), and real-world constraints. Conclusion: The diversification and incorporation of a novel asset class into a portfolio not only mitigates risk but also facilitates the exploitation of opportunities within other investment domains. This research possesses potential utility for stock market participants and organizations engaged in project-based activities that perceive a need for investment within stock markets.