Research Article
Mostafa Sharif; Parisa Shahnazari-Shahrezaei; Meysam Doaei
Abstract
This paper introduces a two-stage stochastic optimization model for portfolio selection, designed to address decision-making uncertainties in the context of the Iranian stock market. The model accounts for a range of disruption scenarios—including economic sanctions, oil price fluctuations, political ...
Read More
This paper introduces a two-stage stochastic optimization model for portfolio selection, designed to address decision-making uncertainties in the context of the Iranian stock market. The model accounts for a range of disruption scenarios—including economic sanctions, oil price fluctuations, political instability, and currency devaluation—enabling dynamic portfolio adjustments to optimize risk-adjusted returns. To manage extreme downside risks, it employs Conditional Value-at-Risk (CVaR) as the risk measure, while simultaneously aiming to maximize expected returns. Compared to traditional mean-variance portfolio optimization, the proposed model demonstrates clear advantages by adapting to uncertain market conditions through scenario-based rebalancing. Sensitivity analysis highlights the model’s responsiveness to critical parameters such as risk aversion, scenario probabilities, and adjustment costs, offering valuable insights into their impact on portfolio performance. The results show that the two-stage model delivers stronger risk management and improved return outcomes than static approaches. Nevertheless, limitations exist, particularly regarding the reliance on accurate scenario probabilities and the assumption of fixed adjustment costs, which may affect real-world applicability. Future research could enhance the model by applying machine learning to refine probability estimates, extending its use to other emerging markets, and integrating more flexible and dynamic cost structures for asset reallocation. The proposed model provides a robust framework for managing investment portfolios in volatile and uncertain environments.
Research Article
Farshid Mehrdoust; Arezou Karimi
Abstract
Precise modeling of financial asset volatility is significant for robust risk management and derivative pricing. Recent scholarly investigations have demonstrated a significant interest in employing stochastic processes with short-term memory for this purpose. Consequently, rigorous ...
Read More
Precise modeling of financial asset volatility is significant for robust risk management and derivative pricing. Recent scholarly investigations have demonstrated a significant interest in employing stochastic processes with short-term memory for this purpose. Consequently, rigorous examination of the existence and uniqueness of solutions for these processes assumes critical importance. This study commences with the precise definition of a fractional operator for $H \in(0, \frac{1}{2})$. Subsequently, the finiteness of the second-order moment of the Itô-Skorokhod integral is meticulously investigated, utilizing the aforementioned operator, specifically within the range of $H \in(0, \frac{1}{2})$. Ultimately, leveraging this moment and rigorously applying Lipschitz and linear growth conditions, and through the application of Gronwall's inequality, the existence and uniqueness of solutions for stochastic differential equations with short-term memory are definitively established.
Research Article
Ehsan Zanganeh; Nafiseh Keshtgar
Abstract
Over recent decades, Iran’s economy has faced significant challenges, including international sanctions, severe exchange rate fluctuations, and high inflation rates, all of which have the potential to drastically alter the trajectory of economic growth. This study investigates the dynamic impacts ...
Read More
Over recent decades, Iran’s economy has faced significant challenges, including international sanctions, severe exchange rate fluctuations, and high inflation rates, all of which have the potential to drastically alter the trajectory of economic growth. This study investigates the dynamic impacts of exchange rate volatility, financial development, trade openness, and inflation on Iran's economic growth over the monthly period from 2011 to 2024, using a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model. This nonlinear approach is adopted due to the limitations of linear models in capturing such complex dynamics. The findings reveal that both exchange rate volatility and financial development exert a negative and statistically significant impact on economic growth, whereas trade openness contributes positively over the long term. Inflation is also found to have a detrimental long-run effect on growth. In the short run, economic growth responds asymmetrically to these variables across different time periods. These results underscore the necessity for policymakers to account for such asymmetric effects when designing and implementing economic policies, especially in contexts affected by currency shocks and sanctions.
Research Article
Yones Esmaeelzade Aghdam; Hamid Mesgarani; Ali Heidarvand
Abstract
Predicting the price of Ethereum remains a significant challenge due to the extreme volatility and nonlinear dynamics inherent in the cryptocurrency market. This study proposes a novel hybrid model that integrates a Gated Recurrent Unit (GRU) with a Transformer Encoder to effectively capture both short-term ...
Read More
Predicting the price of Ethereum remains a significant challenge due to the extreme volatility and nonlinear dynamics inherent in the cryptocurrency market. This study proposes a novel hybrid model that integrates a Gated Recurrent Unit (GRU) with a Transformer Encoder to effectively capture both short-term and long-term temporal dependencies for enhanced Ethereum price forecasting. The model was trained on daily historical data from 2017 to 2023. The dataset, sourced from Yahoo Finance, includes Ethereums open, high, and low prices, along with its trading volume. Additionally, Bitcoins closing price and two technical indicators, On-Balance Volume (OBV) and Average True Range (ATR), were incorporated. Pearson and Spearman correlation analyses confirmed strong interdependencies among the selected features. The model underwent training for 90 epochs, utilizing the Mean Squared Error (MSE) as the loss function and the Adam optimizer. Under identical experimental conditions, the proposed hybrid model significantly outperformed several baseline architectures, including standalone GRU, Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), Transformer Encoder, and CNN–GRU hybrid models. Specifically, the model achieved a Mean Absolute Error (MAE) of 0.007199 (equivalent to $34.03), which is considerably lower than Ethereums average daily price fluctuation of $74.73. These findings demonstrate that the GRU–Transformer Encoder hybrid model is highly effective in extracting intricate patterns from volatile financial time series. Consequently, it can serve as a practical and robust tool for market trend analysis and risk management.
Research Article
Reenu Yadav; Wajahat Ali; Monika Arora
Abstract
This study develops an optimization-based framework for gamification in financial technology (FinTech) to enhance customer loyalty and advance Sustainable Development Goals (SDGs). Data were collected from 33 empirical studies conducted between 2015 and 2025 and analyzed through meta-analysis using the ...
Read More
This study develops an optimization-based framework for gamification in financial technology (FinTech) to enhance customer loyalty and advance Sustainable Development Goals (SDGs). Data were collected from 33 empirical studies conducted between 2015 and 2025 and analyzed through meta-analysis using the Sidik-Jonkman method. The findings demonstrate that gamification features in FinTech platforms significantly strengthen customer loyalty (β = 5.066, p < 0.001), highlighting the central role of engagement mechanisms such as points, rewards, and leaderboards. The residual heterogeneity estimates (τ² = 0.002; I² = 0.604%) reveal low variability across studies, confirming the efficiency and reliability of optimization-driven gamification strategies. Moreover, results indicate that FinTech innovations contribute positively to sustainable development outcomes, with optimization-based gamification mediating between FinTech adoption and SDG achievements. The absence of publication bias (Egger's test: z = -0.543, p = 0.587) further reinforces the robustness of the findings. This study provides strong empirical evidence that integrating gamification with optimization techniques fosters customer loyalty and aligns digital finance practices with global sustainability priorities. It advances theoretical understanding of how digital engagement tools interact with optimization methods to produce measurable social and economic outcomes. It also offers actionable insights for researchers, policymakers, and industry practitioners. The framework opens new pathways for designing customer-centric FinTech solutions that drive business growth and contribute to realizing SDGs.
Research Article
Vesna Lesevic
Abstract
This paper investigates multifactor affine models of the term structure of interest rates, focusing on those that admit closed-form solutions for zero-coupon bond prices. In particular, we study multifactor Vasicek and Cox–Ingersoll–Ross (CIR) models and their hybrid combinations, which integrate ...
Read More
This paper investigates multifactor affine models of the term structure of interest rates, focusing on those that admit closed-form solutions for zero-coupon bond prices. In particular, we study multifactor Vasicek and Cox–Ingersoll–Ross (CIR) models and their hybrid combinations, which integrate Gaussian and square-root dynamics within a single affine framework. By providing a unified analytical treatment, the paper clarifies the economic interpretation of model parameters and explores how they shape the spot and forward rate curves. The hybrid approach enhances the flexibility of term structure modeling, allowing one to capture level, slope, and curvature of the yield curve more accurately than single-class models. These results are directly applicable in practice for yield curve estimation, empirical calibration, risk management, and the pricing of interest rate derivatives.
Research Article
Sureyya Temelli
Abstract
This study investigates the dynamic relationship between stock prices and trading volumes in Borsa İstanbul using the wavelet coherence approach. Employing daily data from major sectoral indices, the analysis captures both time- and frequency-domain interactions between price and volume movements. The ...
Read More
This study investigates the dynamic relationship between stock prices and trading volumes in Borsa İstanbul using the wavelet coherence approach. Employing daily data from major sectoral indices, the analysis captures both time- and frequency-domain interactions between price and volume movements. The methodology enables detection of multi-scale dependencies and shifting lead–lag dynamics across different market phases. The results reveal significant coherence during high-volatility periods, suggesting strong information transmission between price and trading activity. Moreover, the findings indicate that short-term fluctuations are primarily driven by speculative behavior, while long-term linkages reflect fundamental market adjustments. These insights contribute to a deeper understanding of market efficiency and investor behavior in emerging markets. The study provides empirical evidence useful for policymakers, traders, and researchers seeking to interpret complex market structures within a time–frequency framework.
Research Article
Tahar Mohamed Boukadoum; Kamel Boukhetala
Abstract
In the context of financial risk management, predictive modeling under censored data remains a complex challenge. This paper develops and compares two approaches: a traditional log-normal regression model and a hybrid framework combining log-normal regression with an XGBoost-based correction layer. While ...
Read More
In the context of financial risk management, predictive modeling under censored data remains a complex challenge. This paper develops and compares two approaches: a traditional log-normal regression model and a hybrid framework combining log-normal regression with an XGBoost-based correction layer. While the parametric component captures the structured relationships between covariates and claim costs, the machine learning layer adjusts for nonlinear residual structure. Building on this, we introduce a stochastic interpretation of the hybrid estimator by modeling prediction errors as a Gaussian process. We derive a formal variance decomposition, separating model-based and correction-layer uncertainty. To quantify this, we implement both simulation-based estimation and diagnostic tools for residual stationarity and ergodicity. Additionally, we propose a Bayesian stochastic extension by placing priors over model parameters and deriving posterior predictive intervals. A novel contribution of this work is the incorporation of residual dynamics via autoregressive stochastic processes, where residuals from the hybrid model are modeled as AR(1) processes and also as a Diffusion Process. This allows for modeling temporal dependence and improves interpretability of correction structures.
Research Article
Jayanth Athipatla
Abstract
We introduce a novel rough Bergomi (rBergomi) model featuring a variance-driven exponentially weighted moving average (EWMA) time-dependent Hurst parameter $H_t$, fundamentally distinct from recent machine learning and wavelet-based approaches in the literature. Our framework pioneers a unified rough ...
Read More
We introduce a novel rough Bergomi (rBergomi) model featuring a variance-driven exponentially weighted moving average (EWMA) time-dependent Hurst parameter $H_t$, fundamentally distinct from recent machine learning and wavelet-based approaches in the literature. Our framework pioneers a unified rough differential equation (RDE) formulation grounded in rough path theory, where the Hurst parameter dynamically adapts to evolving volatility regimes through a continuous EWMA mechanism tied to instantaneous variance. Unlike discrete model-switching or computationally intensive forecasting methods, our approach provides mathematical tractability while capturing volatility clustering and roughness bursts. We rigorously establish the existence and uniqueness of solutions via rough path theory and derive martingale properties. Empirical validation in diverse asset classes including equities, cryptocurrencies, and commodities demonstrates superior performance in capturing dynamics and out-of-sample pricing accuracy. Our results show significant improvements over traditional constant-Hurst models.
Research Article
Reenu Kumari
Abstract
This study examines the impact of clustered Decision-Making Units (DMUs) in DEA cross-efficiency evaluation, taking into account variables with both positive and negative values. The Range Directional Measure (RDM) model is often used when DMUs have both positive and negative data. However, its application ...
Read More
This study examines the impact of clustered Decision-Making Units (DMUs) in DEA cross-efficiency evaluation, taking into account variables with both positive and negative values. The Range Directional Measure (RDM) model is often used when DMUs have both positive and negative data. However, its application frequently results in clustered DMUs, where multiple units obtain identical RDM cross-efficiency scores, thereby reducing discriminatory power and limiting the reliability of rankings. To overcome this drawback and enhance ranking reliability, we characterize clustered DMUs as scenarios in which identical scores lead to groups of DMUs being termed as ‘clustered’. The necessity of this research lies in improving the robustness of efficiency analysis, especially in situations where both positive and negative data are involved. We then present an algorithm to identify potential clusters and propose a novel clusterless cross-efficiency evaluation method, which restores discrimination and provides more credible performance analysis in decision-making contexts. To demonstrate the practical relevance and advantages of the proposed method, a case study on stock selection in Iran’s stock market portfolio is provided.
Research Article
Somayeh Fallah
Abstract
We develop an option-pricing framework that couples the Heston stochastic volatility model with a fractional Vasicek short-rate process to incorporate long-memory effects in interest rates. Using a regularized semimartingale approximation of fractional Brownian motion and an affine surrogate representation, ...
Read More
We develop an option-pricing framework that couples the Heston stochastic volatility model with a fractional Vasicek short-rate process to incorporate long-memory effects in interest rates. Using a regularized semimartingale approximation of fractional Brownian motion and an affine surrogate representation, we derive a tractable pricing PDE and a semi-closed-form characteristic function suitable for Fourier-based valuation. The numerical implementation employs adaptive integration for fast and accurate pricing, and sensitivity analysis highlights the role of the memory parameter $\alpha$, the smoothing term $\varepsilon$, and the short-rate volatility $\sigma_r$. Empirical calibration to S\&P~500 option data demonstrates that the proposed model improves the fit to market prices relative to the classical Heston model, particularly for longer maturities. These results indicate that persistent interest-rate dynamics can materially influence equity option valuation and motivate further development of fractional interest-rate modelling.
Research Article
Moslem Jamalpour Malekabadi; Ali Zakeri; Amir Hossein Salehi Shayegan
Abstract
In this paper, we propose an approximate solution to a one-dimensional inverse parabolic problem using radial basis functions (RBFs) and the Levenberg–Marquardt (LM) regularization method. This problem involves the backward heat equation. In particular, we first transform the well-known Black–Scholes ...
Read More
In this paper, we propose an approximate solution to a one-dimensional inverse parabolic problem using radial basis functions (RBFs) and the Levenberg–Marquardt (LM) regularization method. This problem involves the backward heat equation. In particular, we first transform the well-known Black–Scholes equation into the heat equation through an appropriate change of variables. The resulting heat equation is then solved using the proposed numerical method.To obtain the approximate solution for the unknown temperature values at the initial time, an optimization problem is formulated to minimize a cost functional. Since the system of equations is ill-conditioned, the LM regularization method is applied. We derive convergence rates for the LM iterates under a Holder stability estimate. Finally, a numerical example is presented to illustrate the method's accuracy and effectiveness.
