This study examines customer retention intention in neobanking environments using a theory-informed explainable machine learning framework. Existing digital banking research typically relies on linear modelling approaches to explain retention behaviour, potentially overlooking nonlinear, value-range-dependent, and interaction-based predictive patterns. Using a publicly available survey of 305 neobank users, this study compares regularized linear models, a partial least squares structural equation modelling (PLS-SEM)-inspired benchmark, and XGBoost (version 3.2.0) under repeated nested cross-validation. SHapley Additive exPlanations (SHAP)-based explainability, SHAP interaction analysis, generalized additive model (GAM) diagnostics, construct-level aggregation, and construct-sensitivity checks are used to interpret model behaviour and assess robustness. The results show that XGBoost substantially outperforms the linear benchmarks, achieving the lowest average RMSE and highest average R2 across 100 out-of-sample test-fold estimates. Trust-related indicators provide the largest share of model-based predictive importance, followed by perceived security and switching costs. SHAP and GAM diagnostics suggest that trust and switching costs may contribute to retention intention in heterogeneous and nonlinear ways, while perceived security displays a more stable positive predictive pattern. Age-related nonlinearities appear weak and should be interpreted cautiously given the young sample profile. The analysis also suggests possible non-additive relationships between trust and perceived security. The study contributes to digital banking and FinTech research by showing how explainable machine learning can complement theory-driven retention models, identify potentially nonlinear predictive patterns, and preserve interpretability. The findings offer practical insight for trust-building, visible security assurance, and retention diagnostics in neobanking contexts.

This study examines the dynamics of technology transfer readiness and financial innovation capability transitions across the expanded BRICS economies, benchmarked against advanced innovation systems through explainable AI. Using a composite Innovation Capability Development-Readiness index (ICDI) constructed through principal component analysis, the paper evaluates the structural conditions enabling knowledge diffusion, industrial upgrading, and financial innovation ecosystem development. A Markov transition framework is employed to analyse how countries evolve across readiness tiers over time, capturing both persistence and mobility in innovation capabilities. The results reveal significant asymmetries in transition probabilities between advanced economies and emerging innovation systems, with several BRICS economies demonstrating gradual upgrading trajectories while others remain structurally locked in lower readiness states. These findings highlight the institutional and policy conditions required to strengthen technology transfer ecosystems. Successful countries in these areas attract foreign investment, participate in global value chains, and profit from technology partnerships. The study contributes to the literature on innovation capability formation and industrial transformation by integrating composite readiness measurement with dynamic transition modelling to inform evidence-based innovation policy.

PurposeThis study investigates the impact of Environmental, Social, Governance (ESG) overall score and its pillars on firm risk and the mediating role of earnings management.Design/methodology/approachThe research applies Generalised Method of Moments (GMM) regression to address endogeneity in a panel of Australian-listed firms from 2014 to 2023.FindingsThe findings reveal that higher ESG scores are associated with lower firm risk, with governance and social pillars exerting the most substantial immediate effects. In contrast, the environmental pillar demonstrates a delayed risk-reducing impact, reflecting long-term benefits rather than short-term volatility reduction. Moreover, the study identifies earnings management as a significant mediator that partially offsets ESG's stabilising effects, highlighting that firms with strong ESG practices are less likely to engage in accrual-based earnings management, thus reducing risk.Practical implicationsThese findings have critical implications for investors, regulators, and policymakers. They underscore the importance of pillar-level ESG evaluation, long-term orientation in environmental assessments and integrating financial transparency into ESG frameworks.Originality/valueThis study contributes to the extant knowledge of ESG overall and the individual pillar effect on firm risk in Australian companies, highlighting the mediating role of earnings management (EM). By identifying earnings management as a partial mediating mechanism, the study extends agency and stakeholder theories beyond direct ESG-firm risk association through the lens of financial reporting behaviour. This integrated framework bridges sustainability and earnings management literatures, offering a more comprehensive theoretical understanding of how ESG performance is related to firm risk.

This study develops an empirical audit framework for assessing the explainability–reliability gap in fraud detection: whether stable model explanations remain consistent with performance-based feature reliance under distribution shift. Using the Bank Account Fraud dataset suite, including a Base dataset and five biased variants, the study examines group-size disparity, fraud-prevalence disparity, separability bias, and temporal shift. Logistic Regression, Linear SVC, and Random Forest are benchmarked using standard classification metrics, followed by cross-variant evaluation with Logistic Regression as the interpretable baseline. SHAP is used to assess explanation stability, while permutation importance measures performance-based feature reliance. The results show that accuracy and ROC-AUC can overstate practical effectiveness under severe class imbalance; notably, Random Forest retained useful discrimination while producing near-zero recall at the evaluated threshold. SHAP feature rankings remained relatively stable across variants, particularly for address-history, identity-similarity, credit-risk, device, and behavioral variables. However, permutation importance revealed weaker and more variable reliance on several SHAP-ranked features. The limited agreement between the two measures indicates a partial explainability–reliability gap. The findings show that explanation stability alone is insufficient for evaluating trustworthy fraud detection models and should be complemented by performance-based validation under biased and shifted deployment conditions.

The development of online banking has brought about an increase in fraudulent operations, which is a major problem for banks. This study delves into the urgent requirement for interpretable, scalable, and top-notch fraud detection systems by using TabNet, an adaptable deep learning framework, on a Kaggle dataset consisting of actual bank transactions in India. Maximizing operational risk management by improving the accuracy of transaction anomaly detection and ensuring regulatory compliance through transparent models is the goal.We utilize a supervised learning pipeline that incorporates the Synthetic Minority Over-sampling Technique (SMOTE) to ensure that classes are balanced. Subsequently, we conduct thorough exploratory data analysis (EDA) to identify patterns of fraud, both during specific times and across behaviors. On this dataset, five different deep learning architectures are tested: DNN, GRU, LSTM, CNN1D, and TabNet. Assessment of predictive performance was carried out using a 3-fold cross-validation framework. With a ROC-AUC of 0.9739 and an accuracy of 97.39%, TabNet considerably outperformed the competition. The method of sparse feature selection used improved interpretability, generalized better on tabular data, and produced fewer false positives and negatives.Critical insights for operational fraud detection systems and a contribution to the broader literature on explainable AI (XAI) in financial decision-making are offered by the findings. Goals 8 and 16 of the Sustainable Development Agenda are supported by this study, which promotes inclusive economic growth and institutional transparency. Supporting strong, policy-compliant, and interpretable decision-support systems, it also offers practical use for real-time implementation in banking infrastructure.

This paper studies the impact of various dimensions of investor sentiment on the stock price response to changes in environmental, social and governance (ESG) ratings. In contrast to most of the existing literature that relies on a single aggregate proxy of sentiment, the present study decomposes investor sentiment into five components such as positive tone, negative tone, risk, volatility and management-related sentiment. The multidimensional design can better pinpoint the channel through which ESG-related information is incorporated into prices. The empirical results show that ESG downgrades are followed by economically meaningful negative cumulative abnormal returns, whereas upgrades generate only weak and short-lived effects. In addition, the pricing response is not conditioned equally by all sentiment dimensions. Positive sentiment is found to be the most important moderating channel. We show that the negative valuation effect of ESG downgrades is significantly stronger in a more optimistic surrounding informational environment, while the other sentiment dimensions play relatively limited roles. Our result is especially pronounced for large firms and for firms with strong pre-event ESG profiles, both of which are likely to attract greater investor attention and to be held more widely by institutional and ESG-oriented investors. The paper contributes to the literature by showing that investor sentiment should not be treated as a single conditioning variable in ESG pricing. Instead, stock price reactions reflect a multidimensional sentiment mechanism in which optimism plays the central role.

This study proposes a fraud detection framework for FinTech that integrates tabular deep learning with relational graph reasoning. We address a key limitation of state-of-the-art tabular fraud models: their assumption that transactions are independent, despite evidence that modern fraud often arises from coordinated infrastructure reuse (e.g., shared devices, email domains, accounts, and addresses). Using the IEEE-CIS Fraud Detection dataset (≈590,540 e-commerce transactions), we develop a hybrid architecture that combines TabNet, chosen for its built-in interpretability on tabular data, with a GraphSAGE encoder trained on a leakage-free heterogeneous entity graph constructed strictly from historical training transactions. Unseen entities in validation and test splits are mapped to dedicated unknown nodes, mirroring deployment conditions and preventing optimistic bias. Class imbalance is handled using square-root class weighting, avoiding synthetic oversampling methods that could distort real infrastructure topology. Empirically, the standalone TabNet baseline achieves ROC–AUC = 0.759 and PR–AUC = 0.149 on a forward-in-time test split. Incorporating relational learning via the proposed TabNet–GraphSAGE fusion improves performance to ROC–AUC = 0.791 and PR–AUC = 0.206, representing a 38% relative gain in PR–AUC and substantially enhancing minority fraud retrieval. The graph encoder alone attains meaningful predictive power, and ablation results confirm that the largest gains emerge from learned message passing and fusion, indicating complementarity between behavioural and infrastructural representations. Overall, results demonstrate that fraud detection is fundamentally a hybrid problem: combining local transaction anomalies with global relational structure yields more operationally relevant ranking quality and improved robustness in high-volume FinTech environments.

In the context of the modern global economy, key drivers of radical change are coming to the fore-transformational transitions to a more complex model-the knowledge economy in its broadest sense. A shift of this magnitude requires new approaches to capacity building for all countries, regardless of their level of development. The emerging imperative provides a wide range of opportunities for developing countries, suggesting a new configuration of the global economic landscape and strategic alliances. The BRICS countries (China, India, Russia, Brazil, UAE, Indonesia, South Africa, Iran, Egypt, Ethiopia) represent a critical mass of players whose actions will determine the direction in which these processes will develop. The article provides a comparative analysis of the innovative and transformational potential of the participants in this block in knowledge-intensive sectors in comparison with each other and with two developed benchmark countries-Austria and Australia. A methodology for calculating a comprehensive index of readiness for the knowledge economy is proposed. The countries under review are ranked on four levels based on the values of this indicator. Their degree of adaptation and readiness for transformational transitions to more complex levels, resource base, educational systems, indicators of patent publication activity, management efficiency, quality of human capital, infrastructure, and global integration capabilities are assessed. Recommendations are presented on policy measures for the development of intellectual economy sectors in the BRICS countries, as well as directions for future research.

Purpose This study aims to examine whether institutional quality moderates the relationship between national intellectual capital and green economic growth. Design/methodology/approach Using annual data for seven ASEAN economies over 2000–2023, the authors use country-specific autoregressive distributed lag models with an error-correction representation to identify long-run cointegrating relationships and short-run adjustments among green economic growth, national intellectual capital, institutional quality and their interaction. Findings The authors compute long-run marginal effects of national intellectual capital at observed levels of institutional quality and identify institutional thresholds at which the effect changes sign. The results indicate systematic, but country-specific, moderation. Brunei, Thailand and Vietnam exhibit stronger contributions of national intellectual capital as institutional quality improves. In Brunei and Vietnam, the marginal effect shifts from negative at low levels of institutional quality to positive at higher levels. Indonesia shows a positive marginal effect across the full range of institutional quality, with only modest attenuation as institutional quality rises. Practical implications The authors’ findings imply that progress toward SDG 8 in ASEAN requires aligning institutional reforms with investments in human capital, research and development, intellectual property services and digital infrastructure, so that knowledge-based capabilities translate into more sustainable productivity and employment. Originality/value This study contributes to the green growth literature by focusing on national intellectual capital as a broad intangible asset base and by providing country-level evidence that its effect on green economic growth depends on institutional quality

This research aims to identify the role of climate finance in sustainability by utilizing climate financing data at the Asian Development Bank (ADB), a major regional MDB working in the most populous, dynamic, and sizeable greenhouse gas emissions areas. We employ the GDP per capita and carbon dioxide (CO2) emissions per capita to proxy for economic and environmental sustainability, respectively, of 39 developing countries financed by ADB from 2016 to 2023. We estimate our research models using various econometric methods that are tailored to the features of our data sample, including DOLS, PCSE, bootstrap panel quantile regression, two-step system GMM approaches, and Common Correlated Effects model. We find a significant impact of ADB climate finance on promoting GDP per capita and reducing CO2 emissions in recipient countries.Nevertheless, this effect varies across the distributional levels of sustainability. Specifically, climate investments are more beneficial to the economic sustainability of lower income countries, while producing greater effects in lowering CO2 emissions in countries with higher emissions.Additionally, we examine the impact of climate adaptation and mitigation funds on sustainability. The results confirm their significant role in enhancing sustainability, with climate adaptation finance having a greater impact than mitigation financing.We also decompose ADB climate finance into sovereign and non-sovereign operations to investigate whether they impact sustainability. Our statistics indicate that both financing operations increase the annual GDP per capita; however, sovereign funds have an inverse effect on reducing CO2 emissions. Our findings thus provide valuable policy implications for MDBs in directing their climate finance to their developing member countries.

This study identifies the dynamics of sentiment contagion between individual and institutional investors in the context of green bond markets across China, Japan, the United States, and the European Union (EU), covering the period from 3 January 2022 to 31 December 2024. Applying a time-varying parameter vector autoregression model, we construct sentiment contagion indicators to capture the evolving interdependencies in the sentiment of the two investor groups. Subsequently, we investigate the impact of these contagion effects on the performance of domestic and cross-border green bond markets by applying exponential general autoregressive conditional heteroscedasticity and quantile-on-quantile regression techniques. Our empirical results reveal consistently high levels of sentiment contagion, with particularly pronounced effects observed in Japan and the EU. The findings underscore the crucial role of sentiment spillovers in shaping green bond markets' performance, although the magnitude and direction of these effects vary across countries and quantiles. Our research findings contribute to the growing literature on sustainable and behavioral finance. It also offers valuable policy implications and investment strategies to green bond regulators and investors across countries.

Green economic growth advocates for a balanced advancement of economic development and environmental sustainability, recognizing the limitations imposed by resource availability and ecological capacity. In many resource-rich developing nations, environmental degradation and inefficient resource use have led to suboptimal green growth outcomes. Amid this challenge, the role of intangible assets, particularly national intellectual capital, in driving green growth remains underexplored. This study examines the moderating effect of institutional quality on the link between national intellectual capital and green economic growth across 19 Asia-Pacific economies from 2000 to 2023. Utilizing dynamic panel estimation techniques, including Dynamic Ordinary Least Squares (DOLS), Fully Modified Ordinary Least Squares (FMOLS), and the Dumitrescu and Hurlin (2012) panel causality method, which account for both cross-sectional dependence and slope heterogeneity, we provide robust empirical insights. Our results reveal that while national intellectual capital and institutional quality individually promote green economic growth, their combined interaction may yield diminishing returns. These findings underscore the importance of aligning institutional frameworks with innovation ecosystems. Accordingly, policy initiatives aimed at sustainable development in the Asia-Pacific region should prioritize strategic knowledge management and support technological progress in the context of the Fourth Industrial Revolution.

The impact of intellectual capital on green innovation has been extensively studied at the firm level. However, the influence of moderating factors on this dynamic at the national level remains underexplored in previous studies. This study examines the role of institutional quality in moderating the relationship between national intellectual capital and green innovation across seventeen Asia–Pacific economies over the last twenty years, starting from 2000. Various techniques are employed to account for cross-sectional dependence and slope homogeneity in panel data analysis, enabling the examination of this relationship over the long and short term. The study also considers the marginal effects of national intellectual capital on green innovation at different degrees of institutional quality. Overall findings indicate that increasing national intellectual capital and institutional quality increases green innovation. Interestingly, the effects of national intellectual capital on green innovation intensify with a greater degree of institutional quality. We also find that enhancing economic growth and the efficient exploitation of natural resources appear to stimulate green innovation in Asia–Pacific economies. Findings imply that policies to improve green innovation should align with traditional economic growth strategies and effectively leverage intangible resources, particularly national intellectual capital. This unique empirical study examines the moderating role of institutional quality in the national intellectual capital–green innovation nexus in Asia–Pacific economies.

This study undertakes a systematic literature review combined with bibliometric analysis to examine how abnormal returns are studied in relation to environmental, social, and governance (ESG) factors, investor sentiment, and dividend policy. Using RStudio version 2025.09.0+387 and VOSviewer version 1.6.20, we conduct a bibliometric study that integrates performance analysis, science mapping, and network analysis. The dataset consists of 532 publications published between 2000 and 2025 and indexed in the Web of Science and Scopus databases. Our results show that scholarly work on abnormal returns is organised around three main thematic areas. First, investor sentiment is closely linked with event study applications, behavioural finance explanations, and sentiment analysis, which underscores the importance of psychological influences in understanding market anomalies. Second, prior studies on dividend policy continue to rely heavily on event study designs to evaluate how markets react to dividend announcements. Third, investor sentiment and dividend policy are connected through their common focus on abnormal returns, which operate as a central conceptual link between these strands of literature. Although interest in behavioural and policy-related determinants of abnormal returns has grown over time, work that explicitly incorporates ESG considerations remains relatively marginal. This peripheral position points to an important gap, suggesting that the dynamic relationships among ESG performance, investor sentiment, dividend decisions, and abnormal returns are still not fully explored. The contribution of this study lies in bringing these elements together by mapping research on event studies while treating ESG performance as a potential market signal that may shape both investor sentiment and corporate financial policy.

This study examines the impact of overall Environmental, Social, and Governance (ESG) performance and its pillars on the default probability of Australian-listed firms. Using a panel dataset spanning 2014 to 2022 and applying the Generalized Method of Moments (GMM) regression, we find that firms with higher ESG scores exhibit a significantly lower likelihood of default. Disaggregating the ESG components reveals that the Environmental and Social pillars have a negative association with default risk, suggesting a risk-mitigating effect. In contrast, the Governance pillar demonstrates a positive relationship with default probability, which may reflect potential greenwashing behavior or an excessive focus on formal governance mechanisms at the expense of operational and financial performance. Furthermore, the analysis identifies trade credit financing (TCF) as a partial mediator in the ESG–default risk nexus, indicating that firms with stronger ESG profiles rely less on external short-term financing, thereby reducing their default risk. These findings provide valuable insights for corporate management, investors, regulators, and policymakers seeking to enhance financial resilience through sustainable practices.

This study describes its objectives by leveraging the significance of the European Union (EU)’s 2050 long-term carbon neutrality strategy to examine the impact of green financing on carbon dioxide (CO2) emissions. The experiments include models based on cointegrated relations, particularly unit root testing, panel cointegration testing, and fully modified ordinary least squares and dynamic ordinary least squares regressions, using EU countries’ panel data covering 2001 to 2019. We first analyze the impact of green finance on CO2 emissions and find an insignificant relationship. When we incorporate biofuel consumption and technological progress, we observe a significant link between green finance and reduced carbon emissions. Our findings suggest that green finance becomes substantially effective in lowering emissions when focusing on the biofuel industry and the transition to low-carbon transportation. The findings provide relevant implications for the EU 2050 long-term strategy for reducing CO2 emissions.

This research investigates the dynamic spillover effects among green bond markets and the impact of investor sentiment on these spillovers. We employ different research methods, including a time-varying parameter vector autoregression, an exponential general autoregressive conditional heteroscedasticity, and a generalized autoregressive conditional heteroskedasticity-mixed data sampling model. Our sample is for twelve international green bond markets from 3 January 2022 to 31 December 2024. Our results evidence the strong correlation between twelve green bond markets, with the United States and China being the net risk receivers and Sweden being the largest net shock transmitter. We also find the varied impact of direct and indirect investor sentiment on the net total directional spillovers. Our research offers fresh contributions to the existing literature in different ways. On the one hand, it adds to the green finance literature by clarifying the dynamic spillovers among leading international green bond markets. On the other hand, it extends behavioral finance research by including direct and indirect investor sentiment in the spillovers of domestic and foreign green bond markets. Our study is also significant to related stakeholders, including investors in their portfolio rebalancing and policymakers in stabilizing green bond markets.

This study examines the relationship between crude oil returns (CRT) and Islamic stock returns (ISR) in BRIC countries during the Global Financial Crisis (GFC), employing wavelet-based comovement analysis and regression models that incorporate both contemporaneous and lagged CRT across 40 cases. The wavelet analysis reveals strong long-term comovement at low frequencies between ISR and CRT during the GFC. Contemporaneous regressions show that increases (decreases) in CRT align with corresponding movements in ISR. Lagged regressions indicate that CRT can predict ISR up to one week ahead for Brazil, Russia, and China, and up to two weeks for India, although the predictive strength weakens beyond this window. These findings challenge the perception that Islamic stocks were immune to the GFC, showing they were affected by global oil market dynamics, albeit with varying degrees of resilience across countries and time horizons.

We conducted a systematic literature review (SLR) by integrating FinTech with the the United Nations' (UN) Sustainable Development Goals (SDGs). By focusing on three dimensions (inclusive finance, economy, and environment), we identify how FinTech may influence sustainable development. We innovate using human- and machine-generated processing to develop themes, making systematic literature reviews more objective and advancing replicability and reproducibility. This study demonstrates the contribution of FinTech in expanding the investment opportunity set by including environmental projects and increasing the diversity and participation rates of savers and lenders. Through this process, FinTech increases its market completeness. Accordingly, FinTech can increase economic growth by achieving higher productivity and sustainable growth through diversification, technological upgrades, entrepreneurship, creativity, and innovation. Additionally, FinTech can accelerate investments in poverty eradication and reduce income inequality. These contributions are aligned with specific SDGs and show that FinTech is an appropriate new technology for financial services.