Research Article
Mohammad Zare
Abstract
Accurate forecasting of asset returns is essential for informed investment decisions and effective portfolio management. This paper explores a hybrid model that combines the Capital Asset Pricing Model (CAPM) with Neural Network Autoregressive (NNAR) models to enhance return predictions. ...
Read More
Accurate forecasting of asset returns is essential for informed investment decisions and effective portfolio management. This paper explores a hybrid model that combines the Capital Asset Pricing Model (CAPM) with Neural Network Autoregressive (NNAR) models to enhance return predictions. While CAPM traditionally estimates expected returns based on market behavior, it has limitations due to its linear assumptions. In contrast, NNAR models excel at capturing complex, nonlinear relationships in financial time series data. Our study integrates NNAR forecasts of market returns into the CAPM framework, hypothesizing that this combined approach will yield superior accuracy, particularly in volatile market conditions. Through empirical analysis, we demonstrate that our hybrid model outperforms traditional CAPM predictions, highlighting the potential of machine learning techniques in asset valuation. The findings provide valuable insights for future research and practical applications in financial forecasting.
Research Article
Mostafa Kebriyayee; Abdolali Basiri; Reza Pourgholi; Rafi Hasani Moghadam
Abstract
The Black-Scholes model is one of the most widely used frameworks for pricing options in financial markets. However, its analytical solutions are often limited to idealized conditions, necessitating the use of numerical methods for more complex scenarios. This study proposes a combined numerical approach ...
Read More
The Black-Scholes model is one of the most widely used frameworks for pricing options in financial markets. However, its analytical solutions are often limited to idealized conditions, necessitating the use of numerical methods for more complex scenarios. This study proposes a combined numerical approach to solve the Black-Scholes equation, specifically focusing on call option pricing in the context of Iran's financial market. The proposed method integrates fully implicit and explicit methods to enhance accuracy and computational efficiency. By applying this approach to historical data from the Iranian options market, we demonstrate its effectiveness in capturing market dynamics and pricing call options under local conditions. The results indicate that the combined numerical method not only provides reliable pricing estimates but also offers insights into the unique characteristics of option trading in emerging markets like Iran. This research contributes to the growing body of literature on numerical methods in financial engineering and provides practical tools for traders and analysts in developing economies.
Research Article
Mehdi Mohammad pour; Majid Zanjirdar; Peyman Ghafari Ashtiani
Abstract
The expansion of communications between active industries and companies in different industry groups on the Tehran Stock Exchange has caused that, in the event of volatility in an industry index, this volatility can spread like a domino to other industry groups and also to other economic sectors, creating ...
Read More
The expansion of communications between active industries and companies in different industry groups on the Tehran Stock Exchange has caused that, in the event of volatility in an industry index, this volatility can spread like a domino to other industry groups and also to other economic sectors, creating systemic risk. Therefore, it is necessary to identify the index of volatile industries, calculate and evaluate the contribution of each of them to the occurrence of systemic risk, the amount of spillover, and the amount of their influence and impact on each other. The purpose of this research is to prioritize the volatility of time series data of 30 industry indices of Tehran Stock Exchange, from 2008 to 2024 using 6 entropy methods, calculate the systemic risk of the growth of each industry index using the conditional value at risk measure ΔCoVaR, and also evaluate the amount of volatility spillover using the TVP-VAR auto-regressive model to predict and prevent the destructive effects of volatility. The research findings show: The highest volatility is related to 8 indices: other mines, communication equipment, agriculture, leather products, coal, petroleum products, chemicals and cement. Also, the highest contagion is to companies active in the coal industry. In addition, the chemical and cement industries can begin to be a systemic risk to the Iranian capital market. Also, a net examination of the spillover effect shows that the growth of the chemical, cement, and communication equipment industries is injecting spillovers into other industries.
Research Article
Mohammad Ali Jafari; Sina Ghasemilo
Abstract
Predicting time series has always been one of the challenges in the financial markets. With the increase in the amount of data, the need to use modern tools instead of classical statistical and time series methods has become clear. In this paper, some deep learning algorithms such as Multilayer Perceptrons ...
Read More
Predicting time series has always been one of the challenges in the financial markets. With the increase in the amount of data, the need to use modern tools instead of classical statistical and time series methods has become clear. In this paper, some deep learning algorithms such as Multilayer Perceptrons (MLPs), Keras Classification, Temporal Fusion Transformer (TFT, developed by Google), Extreme Learning Machine Classification (ELMC) and Propagation Hierarical Learning Network (PHILNet) are used for trading on the foreign exchange market. The efficiency and accuracy of these algorithms are presented. In this order, the EUR/USD data is used as input for the above algorithms.
Research Article
Mohammadreza Rostami; Fatemeh Rasti; Ebrahim Abbasi
Abstract
This study comparatively analyzes two advanced financial risk modeling frameworks: a copula-based Value-at-Risk (VaR) approach and the Multivariate Conditional Autoregressive Value-at-Risk (MCAViaR) model. We assess their effectiveness in capturing risk dynamics across diverse global markets, using daily ...
Read More
This study comparatively analyzes two advanced financial risk modeling frameworks: a copula-based Value-at-Risk (VaR) approach and the Multivariate Conditional Autoregressive Value-at-Risk (MCAViaR) model. We assess their effectiveness in capturing risk dynamics across diverse global markets, using daily log returns from January 1, 2010, to December 31, 2024, for TEDPIX, S&P 500, and BIST 100. This research addresses limitations of traditional linear correlation, especially during market stress.The copula methodology involves two stages: fitting ARMA-GARCH models with Student’s t-distributed innovations for marginal distributions, then employing Gaussian, Student’s t, and Clayton copulas to model inter-market dependence, including tail dependence. MCAViaR, conversely, directly estimates conditional quantiles, adapting to evolving market conditions. Empirical validation is performed through rigorous backtesting, including Kupiec, Christoffersen, and Dynamic Quantile (DQ) tests.Results indicate significant differences. While Student’s t and Clayton copulas effectively capture tail dependence (evidenced by degrees of freedom and positive Clayton parameters), all models—both copula-based and MCAViaR—universally failed the stringent DQ tests across all indices and quantiles. This highlights systematic misspecification in capturing dynamic risk. Despite this, MCAViaR showed a more adaptive nature to sudden market shocks and provided visually more responsive VaR estimates than static copula specifications.The study underscores the necessity of robust, tail-sensitive models for accurate risk assessment in cross-border portfolios. Practical recommendations include adopting Student-t or Clayton copulas, integrating regime-switching mechanisms into MCAViaR, and employing multi-horizon stress testing to enhance dynamic risk management and account for market-specific behaviors.
Research Article
Amir Khorrami; Mahmoud Dehghan Nayeri; Ali Rajabzadeh
Abstract
This study assesses a new simulation-optimization method for credit scoring and bank loan parameter optimization. The proposed approach encompasses data preparation, credit scoring, and simulation-optimization stages. Initially, data regarding bank loans and company financial statements are collected ...
Read More
This study assesses a new simulation-optimization method for credit scoring and bank loan parameter optimization. The proposed approach encompasses data preparation, credit scoring, and simulation-optimization stages. Initially, data regarding bank loans and company financial statements are collected and relevant features are calculated. The Minimum Redundancy Maximum Relevance (MRMR) algorithm selects the most critical features. Subsequently, classification methods including logistic regression (LR), K-nearest neighbor (KNN), artificial neural network (ANN), adaptive boosting (AdaBoost), and random forest (RF) are employed to address the credit scoring problem. These models' performance is evaluated using accuracy, F1-score, and area under curve (AUC) criteria, with the best-performing model selected for subsequent stages. During simulation-optimization, optimal loan features are determined to minimize default rates by treating loan size, interest rate, and repayment period as optimization variables. The memetic algorithm (MA) solves this optimization problem in four cases, utilizing a pre-trained credit scoring model to estimate client default probability. A case study involving 1000 legal clients of an Iranian commercial bank demonstrated that 11 features were selected from 30 defined features for credit scoring. The RF method outperformed other credit scoring models. The simulation-optimization approach reduced default rates from 38% to 20% through decreased loan size and interest rates, coupled with extended repayment periods. These results confirm the method's effectiveness in reducing banking credit risk.
Research Article
Aayush Man Regmi; Samrajya Raj Acharya
Abstract
In an environment marked by financial volatility and rapid economic shifts, reliable forecasts are critical for informed policy-making and strategic financial planning. This study investigates the detailed mathematical exploration followed by its computational performance of time series and deep learning ...
Read More
In an environment marked by financial volatility and rapid economic shifts, reliable forecasts are critical for informed policy-making and strategic financial planning. This study investigates the detailed mathematical exploration followed by its computational performance of time series and deep learning models namely ARIMA, RNN, and TCN applied to foreign assets and liabilities of banking system of Nepal consisting monetary authorities and various depository corporations. Using available primary data, we analyze trends, seasonal patterns, trajectories and their descriptive statistics to capture underlying behaviors. Also, in our work, we have identified the optimal ARIMA order that most effectively captures the linear trend. Empirical study shows that the RNN is able to handle the non-linear patterns which is determined by performance metrices on the training and testing split. TCN being computationally extensive model is not able to capture robust relation of the data due to lack of long-range dependencies and large time windowed dataset for training. Based on our results, the RNN could be used as the most suitable time series forecasting model for the foreign assets and liabilities of Nepal as it enhances the accuracy, minimizes error, and improves effectiveness in contributing to decision-making in banking system.
Research Article
Mahdi Goldani
Abstract
In statistical modeling, prediction and explanation are two fundamental objectives. When the primary goal is forecasting, it is important to account for the inherent uncertainty associated with estimating unknown outcomes. Traditionally, confidence intervals constructed using standard deviations have ...
Read More
In statistical modeling, prediction and explanation are two fundamental objectives. When the primary goal is forecasting, it is important to account for the inherent uncertainty associated with estimating unknown outcomes. Traditionally, confidence intervals constructed using standard deviations have served as a formal means to quantify this uncertainty and evaluate the closeness of predicted values to their true counterparts. This approach reflects an implicit aim to capture the behavioral similarity between observed and estimated values. However, advances in similarity-based approaches present promising alternatives to conventional variance-based techniques, particularly in contexts characterized by large datasets or a high number of explanatory variables. This study aims to investigate which methods—either traditional or similarity-based—are capable of producing narrower confidence intervals under comparable conditions, thereby offering more precise and informative intervals. The dataset utilized in this study consists of U.S. mega-cap companies, comprising 42 firms. Due to the high number of features, interdependencies among predictors are common; therefore, Ridge Regression is applied to address this issue. The research findings indicate that the σ-based method and LCSS exhibit the highest coverage among the analyzed methods, although they produce broader intervals. Conversely, DTW, Hausdorff, and TWED deliver narrower intervals, positioning them as the most accurate methods, despite their medium coverage rates. Ultimately, the trade-off between interval width and coverage underscores the necessity for context-aware decision-making when selecting similarity-based methods for confidence interval estimation in time series analysis.
Research Article
Minou Yari; Mohammad Reza Salehi Rad; Mohammad Bahrani
Abstract
Forecasting financial market volatility has always been a major challenge in economics and financial engineering. In this study, a hybrid approach based on FIGARCH and PLM-GARCH models combined with Long Short-Term Memory (LSTM) neural networks is proposed for modeling financial time series. The analyzed ...
Read More
Forecasting financial market volatility has always been a major challenge in economics and financial engineering. In this study, a hybrid approach based on FIGARCH and PLM-GARCH models combined with Long Short-Term Memory (LSTM) neural networks is proposed for modeling financial time series. The analyzed dataset of the Iran energy index covers October 30, 2016, to January 25, 2023 with 1396 observations. The PLM-GARCH model is capable of identifying long-term dependencies and periodic structures in the conditional variance of time series, while the LSTM network improves prediction accuracy by learning complex and nonlinear patterns. In this approach, the PLM-GARCH model is first used to estimate volatility, and then the residuals from the model are fed as inputs into the LSTM network to extract nonlinear behaviors. Experimental results showed that the combined PLM- GARCH-LSTM model (RMSE = 0.00209, MAPE ≈ 5.1%) outperforms the FIGARCH-LSTM model (RMSE = 0.00224, MAPE ≈ 5.8%) and significantly improves prediction accuracy. These find- ings suggest that combining econometric periodic methods with deep learning can be a powerful tool for forecasting financial volatility.
Research Article
Amir Mohsen Moradi; Mohsen Mehrara; Mahdieh Tahmasebi
Abstract
This paper estimates systematic risk in Iran’s foreign exchange market using a stochastic volatility model, analyzing five distinct episodes shaped by varying economic and political conditions. By tracing the evolution of volatility dynamics across these episodes, we reveal critical shifts in market ...
Read More
This paper estimates systematic risk in Iran’s foreign exchange market using a stochastic volatility model, analyzing five distinct episodes shaped by varying economic and political conditions. By tracing the evolution of volatility dynamics across these episodes, we reveal critical shifts in market behavior under different risk regimes. Our results show that during low-risk episodes, volatility shocks exhibit high persistence, causing market disturbances to linger. In contrast, as systematic risk intensifies, volatility shocks dissipate more rapidly—yet this reduced persistence coincides with a marked rise in average volatility. We identify three particularly turbulent episodes in the past seven years, each characterized by exceptionally high levels of systematic risk. Strikingly, both the mean and variance of volatility increased during these high-risk periods, signaling not only heightened instability but also deeper Knightian uncertainty. These findings carry significant policy implications: when direct reduction of volatility proves challenging, policymakers should prioritize reducing the volatility of volatility to mitigate uncertainty and stabilize expectations. Notably, our analysis indicates that a 1% reduction in volatility corresponds to a 1.7% decline in the variance of daily exchange rate returns, underscoring the leverage policymakers have over market uncertainty.
Research Article
Nooshin Hakamipour
Abstract
Reliability assessment, vital in high-stakes engineering, often employs the stress-strength model. However, traditional models frequently assume independence between stress and strength, an assumption that can lead to inaccurate reliability estimates when dependence exists due to real-world factors. ...
Read More
Reliability assessment, vital in high-stakes engineering, often employs the stress-strength model. However, traditional models frequently assume independence between stress and strength, an assumption that can lead to inaccurate reliability estimates when dependence exists due to real-world factors. To address this, the current study proposes a dependent stress-strength model using copula theory, which flexibly models dependence by separating marginal and joint distributions. Four copula families—Farlie-Gumbel-Morgenstern, Ali-Mikhail-Haq, Gumbel's bivariate exponential, and Gumbel-Hougaard—are investigated for their ability to capture diverse dependency patterns. The Inverse Lomax distribution is utilized for both stress and strength marginals due to its suitability for heavy-tailed reliability data. The copula dependence parameter θ is estimated via conditional likelihood and Blomqvist's beta-based method of moments. The asymptotic distributions of these estimators are derived, and their performance is evaluated through extensive simulations. The research thoroughly examines how system reliability R changes with $\theta$ across various model configurations. Findings indicate that the Gumbel-Hougaard copula demonstrates the highest sensitivity of $R$ to θ, effectively capturing a wide range of dependency strengths. This paper highlights the critical need to incorporate dependence in stress-strength models and offers practical guidance for copula selection, thereby enhancing the accuracy and robustness of reliability predictions in complex engineering systems. A practical examination of a real dataset is conducted to demonstrate the concept.
Research Article
Sayyede Elnaz Afzaliyan Boroujeni; Abdolmajid Abdolbaghi Ataabadi; Naser Khani
Abstract
The present study aims to assess the impact of implied volatility (IV) extracted from call option prices on abnormal stock returns. IV, as a critical market volatility index, plays an essential role in explaining investor behavior. The Black-Scholes model was used to extract IV, applying Brent’s ...
Read More
The present study aims to assess the impact of implied volatility (IV) extracted from call option prices on abnormal stock returns. IV, as a critical market volatility index, plays an essential role in explaining investor behavior. The Black-Scholes model was used to extract IV, applying Brent’s method due to the absence of an explicit closed-form solution. In addition, daily call option trading data from the Tehran Stock Exchange (TSE) were utilized during 2016-24. Further, quantile multivariate regression, along with wild bootstrap resampling (1,000 repetitions), was employed for model estimation. Abnormal returns (ARs) were significantly associated with IV, illiquidity (ILLIQ), daily stock returns (RETs), and bid-ask spreads (SPREAD). However, ARs were negatively correlated with the logarithm of firm size (LogSIZE), historical option volatility (σOption), logarithm of book-to-market (LogBM), implied volatility delta (ΔIV), and idiosyncratic volatility (IDVOL). The historical stock volatility (σStock) and options-to-stocks volume ratio (O/S) demonstrated no significant association with ARs. The results highlighted the predictive power of IV and ΔIV for future price movements. The study recommends market participants and portfolio managers to incorporate the above-mentioned metric into investment decision-making processes.
