Search Results (3,783)

This study addresses discontinuity prediction in income reporting within the Chilean student loan program, a critical event for credit risk management. Although the literature has incorporated machine learning models to anticipate non-compliance behavior, a gap remains in the development of methodologically robust evaluations that integrate nonlinear imputation, imbalance correction, and repeated validation across multiple partitions. To address this need, a complete pipeline was implemented on a dataset of 22,303 records, including MissForest imputation, SMOTE-based balancing, and a comparative assessment of a biologically inspired Deep Neural Network (DNN) and a Random Forest (RF) classifier used as a classical baseline model, evaluated across 35 stratified partitions. The results show that the bioinspired DNN, as the primary focus of this study, consistently outperforms the RF in metrics such as AUC (0.9991 vs 0.9709), F1-score (0.9966 vs 0.9497), and agreement measures, while also exhibiting lower variability across partitions. The interpretability analysis indicates that financial variables account for the greatest influence on predictions, whereas demographic variables contribute minimally. The study provides a replicable and robust methodology aligned with risk analysis practices in student credit contexts. Full article

This study constructs an urban energy consumption structure transformation (UECST) index and utilizes a double machine learning model to investigate the impact and underlying mechanisms of green finance policies on this transformation. Based on panel data from 281 prefecture-level cities in China from 2010 to 2022, we find that green finance policies significantly promote the UECST. This finding holds after a series of robustness checks and endogeneity tests. Furthermore, our analysis reveals that these policies facilitate the transition not only through direct financial support but also indirectly by driving green technological innovation, enhancing green economic efficiency, and promoting industrial upgrading. The positive impact is more substantial in central cities, transportation hubs, non-resource-based cities, non-old industrial bases, and key environmental protection cities. By providing empirical evidence and policy insights, this study contributes to optimizing green finance policy design and addressing specific bottlenecks in energy transition, thereby supporting the achievement of the “Beautiful China” development goal. Full article

This study examines the interrelated challenges of climate change and energy poverty across two distinct industrial regions: Borsod-Abaúj-Zemplén in Hungary and Zarqa in Jordan. Both areas face unemployment and low-income levels, as well as environmental legacies of industrial activity; however, they differ significantly in their energy policies and infrastructure development. Using 2025 survey data, we develop indices of energy poverty, financial poverty, and climate perceptions, aligned with OECD guidelines. Regression analysis indicates that the model accounts for approximately 40% of the variance in energy poverty. Notably, heightened perceptions of climate change are associated with increased reports of energy hardship, suggesting that economically deprived households possess greater climate risk awareness. Resilience capacities, including adaptive skills, income stability, and community support, are found to substantially mitigate energy poverty. Income and employment status also play protective roles, underscoring the importance of economic resources. The impact of financial poverty varies markedly, being negligible in Hungary but severe in Jordan due to structural and infrastructural constraints. Our findings underscore the need for tailored, inclusive policy interventions that emphasize energy efficiency and retrofitting in Hungary and promote financial support and the adoption of renewable energy in Jordan. Integrating principles of energy justice into climate resilience strategies is crucial for promoting equitable and sustainable energy transitions, mitigating local vulnerabilities, and enhancing overall household resilience. Full article

This study presents a hybrid analytical framework that enhances project selection by achieving reasonable predictive accuracy through the integration of expert judgment and modern artificial intelligence (AI) techniques. Using an enterprise-level dataset of 10,000 completed software projects with verified real-world statistical characteristics, we develop a three-step architecture for intelligent decision support. First, we introduce an extended Analytic Hierarchy Process (AHP) that incorporates organizational learning patterns to compute expert-validated criteria weights with a consistent level of reliability ($CR=0.04$), and Linear Programming is used for portfolio optimization. Second, we propose a machine learning architecture that integrates expert knowledge derived from AHP into models such as Transformers, TabNet, and Neural Oblivious Decision Ensembles through mechanisms including attention modulation, split criterion weighting, and differentiable tree regularization. Third, the hybrid AHP-Stacking classifier generates a meta-ensemble that adaptively balances expert-derived information with data-driven patterns. The analysis shows that the model achieves 97.5% accuracy, a 96.9% F1-score, and a 0.989 AUC-ROC, representing a 25% improvement compared to baseline methods. The framework also indicates a projected 68.2% improvement in portfolio value (estimated incremental value of USD 83.5 M) based on post factum financial results from the enterprise’s ventures.This study is evaluated retrospectively using data from a single enterprise, and while the results demonstrate strong robustness, generalizability to other organizational contexts requires further validation. This research contributes a structured approach to hybrid intelligent systems and demonstrates that combining expert knowledge with machine learning can provide reliable, transparent, and high-performing decision-support capabilities for project portfolio management. Full article

The main purpose of this study was to conduct a bibliometric analysis of scientific knowledge and trends in modern finance. To this end, the analysis was based on the keywords: “finance”, “automation”, and “ESG”. The analysis aimed to provide theoretical insights into the economic and financial implications of automation and its role in achieving ESG objectives. From a methodological standpoint, bibliometric research was conducted on 21 September 2025. It involved analysing a total of 16,500 scientific articles published between 1974 and 2026 in two databases: The Web of Science Core Collection and Scopus. The Bibliometrix R 5.2.0 version tool was used to generate visualisations. Thematic mapping, three-field plotting, keyword mapping, and clustering were the main methods used to analyse the associations between finance, automation, and ESG principles. The study’s results showed an average annual increase in publications of approximately 3.80% and 2.50%, respectively, while international collaborations between researchers have become increasingly prominent in recent years. At the same time, the co-occurrence network analysis identified five key thematic clusters in the Web of Science Core Collection and three in Scopus. From a comparative perspective, these clusters highlight the most significant connections between environmental, social, and governance (ESG) performance, corporate social responsibility (CSR) impact, financial performance, economic growth, sustainable development, and the implications of the automation process. From a bibliometric point of view, this research contributes to a better understanding of the multiple digital transformations specific to the current financial framework, generating possible future research directions on the significant role of automation in financial, environmental, and social performance. Furthermore, automation is a critical component of the digital future of finance. Analysing and investigating the causal relationships between automation and Environmental, Social, and Governance (ESG) principles will necessitate new areas of study within the financial sphere. Full article

Graph anomaly detection (GAD) is pivotal for security-critical applications like cybersecurity and financial fraud detection. While reconstruction-based Graph Neural Networks (GNNs) are prevalent, their efficacy is often compromised by two phenomena: (1) anomaly overfitting, where expressive models capture anomalous patterns, and (2) homophily-induced attenuation, where message passing smooths localized anomaly cues. This paper proposes CLEAN-GAD, a contamination-aware framework that mitigates anomaly influence during training through multi-view robust learning. Specifically, we develop a contrastive augmentation module that utilizes local inconsistency scores to identify and suppress pseudo-anomalous nodes and edges, thereby yielding a purified augmented view. To capture diverse anomaly signals, a frequency-adaptive encoder with dual-pass channels is designed to integrate low- and high-frequency information. Furthermore, we introduce a distribution-separation regularizer and cross-view alignment to stabilize learning and resolve view shifts. Theoretical analysis confirms that reducing the contamination ratio $\rho$ expands the reconstruction-risk gap between normal and anomalous nodes, inherently boosting detection performance. Extensive experiments on multiple benchmark datasets from various domains demonstrate the superior anomaly detection performance of CLEAN-GAD. Full article

The research examines organisational leadership styles in the transition to the circular water economy using explanatory quantitative methods, combining semantic normalisation of structured survey responses and latent class analysis. One hundred and fifty organisations from the water sector in Lima, Trujillo, and Cajamarca participated and received a previously validated 30-item Likert-type questionnaire (α = 0.97). The nine analytical domains were developed resources, leadership, culture, technological capabilities, rivalry, suppliers, regulatory framework/support, implementation, and results to discover different organisational configurations. The ideal model identified eight latent classes that are grouped into four organisational archetypes: established leaders, aspirants with regulatory deficits, environment-focused with medium execution, and structural laggards. The findings reveal that circular implementation and results depend more on the articulation between organisational culture, strategic leadership, and regulatory framework than on the availability of technical or financial resources. In addition, great interorganizational heterogeneity was found, which challenged homogeneous public policies and requires differentiated strategies according to the level of circularity in which each organisation finds itself. The research provides empirical evidence to operationalise water transition indicators within the framework of SDG 6 and SDG 12, developing a robust taxonomy to track institutional progress toward water sustainability. Full article

This article synthesizes the impact of parental financial socialization on an individual’s financial behavior. To better understand the role of parental financial socialization, 219 peer-reviewed articles from the Scopus database were analysed. A combination of bibliometric and thematic analysis was used, resulting in four major themes: (1) Mechanisms of parental and family financial socialization, (2) Financial outcomes from family financial socialization, (3) Psychological Mediators of Socialization Effects, and (4) Socio-Cultural and Institutional Contexts as Moderators. Findings of this study reveal that parental modeling, communication, psychology, socio-cultural, and institutional context are key mechanisms in the development of financial norms and competencies. The study also confirms the relevance of the Social Learning Theory, Family Systems Theory, Theory of Planned Behavior, Financial Capability Theory, and Life Course Perspective Theory. The contributions of this study include the development of a multi-level model that identifies family, psychological, and institutional determinants of financial behavior and proposes areas for future research in different cross-cultural contexts. From a practical perspective, this study highlights the importance of integrating the factors mentioned above into policy interventions by regulators and all stakeholders. Full article

This paper investigates how monetary and ESG-related risks—especially those stemming from asymmetric policy transmission across Eurozone economies—have evolved over time, with a focus on the post-COVID-19 era. Using a mixed-method bibliometric analysis of 216 peer-reviewed articles (1996–2025), it maps thematic developments in monetary governance and sustainability discourse. Findings reveal a post-2020 surge in scholarly engagement, marked by a decisive shift: ESG risks, once peripheral, are now central to discussions of macro-financial stability and institutional resilience. This thematic realignment aligns with major EU regulatory milestones (e.g., SFDR, EU Taxonomy, CSRD), signaling a structural transformation in EU governance. The study concludes that the convergence of monetary asymmetry and ESG integration represents a new frontier in economic policy and academic inquiry, raising critical questions about institutional convergence, regulatory capacity, and sustainability-informed monetary frameworks in post-crisis Europe. Full article

The shift toward a circular economy (CE) holds a central position in solving, globally, the long-standing environmental degradation and resource scarcity. Innovative sustainable processes and products lie at the core of such a shift, but they often face challenges associated with technological, organizational, financial, and regulatory paradigms. To date, two leading facilitators have been identified: sophisticated digital technologies, such as Artificial Intelligence, the Internet of Things, and Big Data, together with the collaborative creation of value among diverse stakeholders. Although the implications of each of these enablers on sustainability are known to some extent, little is understood about how their interrelatedness can counteract implementation barriers and drive innovation. The systematic thematic literature review examines how organizations utilize digital technologies and stakeholder engagement to facilitate green innovation in Circular Economy (CE) systems. Based on Stakeholder Theory, the Technology-Organization-Environment (TOE) framework, and the Resource-Based View (RBV), this review examines how organizations leverage digital technologies and stakeholder engagement to foster green innovation within circular economy systems. Following the PRISMA 2020 guidelines, a structured search was conducted in Scopus and Web of Science, covering peer-reviewed journal articles published in English between 2013 and 2024. Using predefined inclusion and exclusion criteria, 84 studies were retained for analysis from an initial pool of 850 records. The review integrates findings from five thematic areas: collaborative innovation among stakeholders, the use of digital technology to advance sustainability, challenges associated with adopting circular-economy values, linkages between technology and stakeholders, and the consequences of innovation. The findings suggest that collaboration between diverse stakeholders, combined with integration with digital technologies, provides a synergistic approach to maximizing innovation outcomes, overcoming implementation challenges, and diffusing circular practice. Skillfully crafted initiatives augment organizational capacities, foster collaborative actions, and advance sustainability initiatives. Despite providing a comprehensive synthesis of existing research, this review is limited by its reliance on secondary data. A qualitative quality appraisal was conducted to support the interpretation of findings. This review was not registered and received no external funding. Future research should conduct empirical analyses of these relationships and develop inclusive frameworks to guide initiatives emerging from collaborative and digital platforms across a wide range of sectors within the circular economy. Full article

From adaptation to building effective resilience to climate change is critical for transforming West and Central Africa (WCA) agricultural system. Climate-Smart Agriculture (CSA) is an approach initiated by leading international organizations to ensure food security, increased adaptation to climate change and mitigation. Its application spans from innovative policies, practices, technologies, innovations and financing. However, CSA initiatives lack scientific-based assessment prior to implementation to ensure their effectiveness. To fill this gap, future interventions should not only be assessed using rigorous methodology but should also be built on lessons learned from previous initiatives. Although there are a lot of climate related agricultural initiatives in WCA, most of them have not been analyzed through a CSA lens and criteria to capitalize on their experiences to improve future interventions. In this study we mapped previous climate-related initiatives in WCA, highlighted their gaps and lessons learned to accelerate the implementation of CSA in the region. The study covered 20 countries in WCA: Benin, Burkina Faso, Cameroon, Cape Verde, Central African Republic, Chad, Côte d’Ivoire, Congo, Gabon, Gambia, Ghana, Guinea, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, Togo. CSA initiatives were reviewed using a three-steps methodology: (i) national data collection, (ii) regional validation of the national database, (iii) data analysis including spatial mapping. Data was collected from the websites of international, regional and national organizations working in the field of agricultural development in the region. Each initiative was analyzed using a multicriteria analysis based on CSA principles. A total of 1629 CSA related initiatives were identified in WCA. Over 75% of them were in the form of projects/programs with more of a focus on the first CSA pillar (productivity and food security), followed by adaptation. The mitigation pillar is less covered by the initiatives. Animal production, fisheries, access to markets, and energy are poorly included. More than half of these initiatives have already been completed, calling for more new initiatives in the region. Women benefit very little from the implementation of the identified CSA initiatives, despite the substantial role they play in agriculture. CSA initiatives mainly received funding from technical and financial partners and development partners (45%), banks (22%), and international climate financing mechanisms (20%). Most of them were implemented by government institutions (48%) and development partners (23%). In total, more than 600 billion EUR have been disbursed to implement 83 of the 1629 initiatives identified. These initiatives contributed to reclaiming and/or rehabilitating almost 2 million ha of agricultural land in all countries between 2015 and 2025. Future initiatives should ensure the consideration of the three CSA pillars right from their formulation to the implementation. These initiatives should consider investing in mixed production systems like crop-animal-fisheries. Activities should be built around CSA innovation platforms to encourage networking among actors for more sustainability. Full article

In the grand context of the global convergence of the “dual-carbon” strategy and the digital economy, the underlying mechanisms by which corporate fintech impacts ESG performance remain a “black box” waiting to be explored. To this end, this study reveals the path by which corporate fintech unlocks ESG performance by constructing a theoretical framework that integrates resources, technology and governance. Based on data from Chinese A-share listed companies from 2011 to 2023, we found that corporate fintech can significantly improve ESG performance. Its core mechanism is to optimize resource allocation by alleviating financing constraints, promote green innovation-driven technological upgrades, and reduce agency costs to improve internal governance. Heterogeneity analysis further reveals that this effect is particularly prominent in companies with financial difficulties or high proportions of independent directors, and areas with weak institutional environments, highlighting the catalytic role of corporate fintech in specific situations. This study not only provides micro-mechanism evidence for digital technology to empower the sustainable development of enterprises but also offers important policy implications for emerging markets to leverage fintech to make up for institutional shortcomings and promote green transformation. Full article

This paper develops a comprehensive macro stress-testing (MST) framework to evaluate the resilience of Saudi Arabia’s financial sector against systemic risk over the period 2010–2025. The approach integrates macro financial linkages, credit risk modeling, and scenario analysis to simulate the impact of severe but plausible shocks on capital adequacy ratios (CAR) and capital shortfalls. Using Saudi macroeconomic data, the study demonstrates that GDP growth and oil price fluctuations are dominant drivers of systemic risk, while inflation and unemployment exert significant but secondary effects. Under severe adverse conditions, the banking sector’s aggregate CAR declines to 9.6%, requiring an estimated capital injection of 3.7% of GDP. The findings underscore the strength of Saudi Arabia’s financial buffers, while emphasizing the importance of dynamic capital buffer calibration, sectoral diversification, and cross-border macroprudential coordination within the GCC. Policy recommendations are provided to enhance stress-testing governance and fiscal and financial alignment. The findings highlight the importance of dynamic counter-cyclical capital buffers, sectoral diversification, liquidity resilience, and enhanced fiscal–financial coordination. Policy recommendations are provided to guide SAMA and the Financial Stability Council in capital planning, stress-test governance, and macroprudential policy design. Full article

Small detector arrays, which are designed to record relatively small EAS with energy in the ‘knee’ region, are often equipped with clocks that measure the time difference between fast signals from several detectors, as well as spectrometric channels that provide the amplitudes of these signals. When analyzing them to determine the angles of arrival and the size of the registered showers, it is important to take into account uncertainties, i.e., the dispersion of measured time differences and shower size relative to the ‘true’ values, which are unknown in the actual situation. Analyses of these spreads are essentially only possible on the basis of correctly performed simulation calculations that take into account all possible stochastic processes in the development of showers in the atmosphere. In this paper, we present a simulation-based analysis using the CORSIKA program of a small EAS array model consisting of four charged particle detectors. We demonstrate the potential offered by ideal timing and how we can infer the energy of the primary particle by analyzing signal amplitudes. The analysis shows that the costs, not only financial, of introducing timing and shower spectrometry are not worth the potential physical gains that we can achieve by using them to analyse small showers. Full article

Aim: Stock price prediction remains a highly challenging task due to the complex and nonlinear nature of financial time series data. While deep learning (DL) has shown promise in capturing these nonlinear patterns, its effectiveness is often hindered by the low signal-to-noise ratio inherent in market data. This study aims to enhance the stock predictive performance and trading outcomes by integrating Singular Spectrum Analysis (SSA) with deep learning models for stock price forecasting and strategy development on the Australian Securities Exchange (ASX)50 index. Method: The proposed framework begins by applying SSA to decompose raw stock price time series into interpretable components, effectively isolating meaningful trends and eliminating noise. The denoised sequences are then used to train a suite of deep learning architectures, including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and hybrid CNN-LSTM models. These models are evaluated based on their forecasting accuracy and the profitability of the trading strategies derived from their predictions. Results: Experimental results demonstrated that the SSA-DL framework significantly improved the prediction accuracy and trading performance compared to baseline DL models trained on raw data. The best-performing model, SSA-CNN-LSTM, achieved a Sharpe Ratio of 1.88 and a return on investment (ROI) of 67%, indicating robust risk-adjusted returns and effective exploitation of the underlying market conditions. Conclusions: The integration of Singular Spectrum Analysis with deep learning offers a powerful approach to stock price prediction in noisy financial environments. By denoising input data prior to model training, the SSA-DL framework enhanced signal clarity, improved forecast reliability, and enabled the construction of profitable trading strategies. These findings suggested a strong potential for SSA-based preprocessing in financial time series modeling. Full article