This study develops a novel deep learning framework for bank distress prediction, designed to overcome the limitations of static network analysis and to enhance model interpretability. We propose a Spatio-Temporal Attention Network that uniquely captures the time-varying nature of systemic risk. Methodologically, it introduces two key innovations: (1) a dynamic interbank network whose connection weights are adjusted by the volatility of the Shanghai Interbank Offered Rate (SHIBOR), reflecting real-time market liquidity changes; and (2) a dual spatio-temporal attention mechanism that identifies critical time steps and pivotal contagion pathways leading to a distress event. Empirical results demonstrate that the model significantly outperforms traditional benchmarks across key metrics including accuracy and F1-score. Most critically, the architecture proves exceptionally effective at reducing Type II errors, substantially minimizing the failure to identify at-risk banks. The model also offers high interpretability, with attention weights visualizing intuitive risk evolution patterns. We conclude that incorporating dynamic, liquidity-adjusted networks is crucial for superior predictive performance in systemic risk modeling.

Periods of market crisis are often accompanied by heightened fear and information overload, which can induce information avoidance behaviors such as the ostrich effect. While prior research has documented investors’ tendency to avoid unfavorable information, little is known about how different information sources—and trust in those sources—jointly shape such behavior under extreme uncertainty. Drawing on Granular Interaction Thinking Theory (GITT) and employing Bayesian Mindsponge Framework (BMF) analytics, this study examines how investors’ regular securities-related information sources is associated with the ostrich effect during the 2022 market downturn in China, and how these associations are conditioned by trust. Using survey data from 1,451 Chinese individual stock investors, we model investors’ recalled frequency of temporarily disengaging from stock investing as an indicator of information avoidance. The results show that regularly consulting professional sources, financial newspapers, and online forums is associated with information avoidance, whereas reliance on personal relationships and company disclosures is not. Importantly, trust moderates these relationships in distinct ways. Higher trust in professional sources is associated with reduced information avoidance, while higher trust in financial newspapers and online forums amplifies avoidance behavior. Among all sources, the interaction between trust and information referral is strongest for financial newspapers. These findings suggest that trust does not uniformly mitigate fear-driven avoidance. Instead, when combined with high-entropy information sources, trust can exacerbate cognitive and emotional strain, increasing investors’ propensity to disengage. By highlighting the joint roles of informational entropy and trust, this study advances behavioral finance research and offers practical insights for investors, policymakers, and regulators seeking to improve decision-making resilience during periods of market crisis.

This study proposes an innovative CPO-VMD-PConv-Informer framework to forecast the KBW Nasdaq Financial Technology Index (KFTX). The framework comprehensively incorporates the effects of eight representative uncertainty indicators on KFTX price predictions, including the Economic Policy Uncertainty Index (EPU) and the Geopolitical Risk Index (GPR). The empirical findings are as follows: (1) The proposed CPO-VMD-PConv-Informer framework demonstrates superior predictive performance across the entire sample period, achieving R² values of 0.9681 and 0.9757, significantly outperforming other commonly used traditional machine learning and deep learning models. (2) By integrating VMD decomposition and CPO optimization, the model effectively enhances its adaptability to extreme market volatility, maintaining stable predictive accuracy even under structural shocks such as the COVID-19 outbreak in 2020. (3) Robustness tests show that the proposed model consistently delivers strong predictive performance across different training-testing data splits (9:1, 8:2, and 6:4), with the MAPE remaining below 2%. These findings provide methodological advancements for forecasting in the KFTX market, offering both theoretical value and practical significance.

This study develops a novel deep learning framework for bank distress prediction, designed to overcome the limitations of static network analysis and to enhance model interpretability. We propose a Spatio-Temporal Attention Network that uniquely captures the time-varying nature of systemic risk. Methodologically, it introduces two key innovations: (1) a dynamic interbank network whose connection weights are adjusted by the volatility of the Shanghai Interbank Offered Rate (SHIBOR), reflecting real-time market liquidity changes; and (2) a dual spatio-temporal attention mechanism that identifies critical time steps and pivotal contagion pathways leading to a distress event. Empirical results demonstrate that the model significantly outperforms traditional benchmarks across key metrics including accuracy and F1-score. Most critically, the architecture proves exceptionally effective at reducing Type II errors, substantially minimizing the failure to identify at-risk banks. The model also offers high interpretability, with attention weights visualizing intuitive risk evolution patterns. We conclude that incorporating dynamic, liquidity-adjusted networks is crucial for superior predictive performance in systemic risk modeling.

In the realm of financial stability, accurate credit default discrimination models are crucial for policy-making and risk management. This paper introduces a robust model that enhances credit default discrimination through a sophisticated integration of a filter-wrapper feature selection strategy, instance selection, and an updated version of majority voting. We present a novel approach that combines individual and ensemble classifiers, rigorously tested on datasets from Chinese listed companies and the German credit market. The results highlight significant improvements over traditional models, offering policymakers and financial institutions a more reliable tool for assessing credit risks. The paper not only demonstrates the effectiveness of our model through extensive comparisons but also discusses its implications for regulatory practices and the potential for adoption in broader financial applications.

Traditional decision tree models, which rely exclusively on numerical variables, often encounter difficulties in handling high-dimensional data and fail to effectively incorporate textual information. To address these limitations, we propose the Integrated Multivariate Segmentation Tree (IMST), a comprehensive framework designed to enhance credit evaluation for small and medium-sized enterprises (SMEs) by integrating financial data with textual sources. The methodology comprises three core stages: (1) transforming textual data into numerical matrices through matrix factorization; (2) selecting salient financial features using Lasso regression; and (3) constructing a multivariate segmentation tree based on the Gini index or Entropy, with weakest-link pruning applied to regulate model complexity. Experimental results derived from a dataset of 1,428 Chinese SMEs demonstrate that IMST achieves an accuracy of 88.9%, surpassing baseline decision trees (87.4%) as well as conventional models such as logistic regression and support vector machines (SVM). Furthermore, the proposed model exhibits superior interpretability and computational efficiency, featuring a more streamlined architecture and enhanced risk detection capabilities.

Assessment in vocational education institutions is frequently organized around performance metrics—graduation rates, employment outcomes, and satisfaction scores—gathered too tardily to avert institutional dysfunction. In increasingly unstable policy situations, these models have become precarious: they quantify collapse more frequently than they avert it. This paper presents fuse logic as an innovative mechanism for risk-responsive governance in technical and vocational education and training (TVET). Utilizing systems control theory and the analogy of circuit breakers, fuse logic is a threshold-sensitive, dynamically activated assessment paradigm designed to disconnect institutional activities prior to complete failure. The research formulates a four-stage model—situational sensing, threshold definition, fuse activation, and adaptive reconfiguration—and implements it in a simulated scenario reflecting Chinese TVET trends. When critical metrics surpass risk thresholds (e.g., dropout rate, employment mismatch), fuse logic triggers systematic program shutdowns, stakeholder consultations, and conditional reintegration procedures.This study's contribution is in redefining evaluation from measurement to protection. It advocates a governance framework that permits temporary disconnection to maintain system integrity. Fuse logic enhances conventional quality assurance frameworks by providing an integrated, failure-tolerant layer of organizational resilience. The report concludes with a discussion on transferability, ethical considerations, and prospective avenues for implementation across varied educational systems.

We propose a novel dual-screening classification framework for fund selection designed to align statistical objectives with investor goals. Testing on the Chinese mutual funds market, a Gradient Boosting model implementing our framework generates a statistically and economically significant 14.65% annual risk-adjusted alpha, substantially outperforming identical models trained under a standard regression framework. Feature importance analysis confirms that fund-level momentum and flows are the most significant predictors of performance in this market. Our findings provide a robust and practical framework for active management, demonstrating that modelling both upside potential and downside risk is critical for superior performance.

We construct the China Crude Oil Volatility Index (CNOVX)—the first model-free, optionimplied measure of forward-looking oil price risk for China—using INE crude oil options from 2021 to 2024 and an adapted CBOE methodology that accounts for sparse strike availability via smooth interpolation and extrapolation. Our results show that CNOVX increases with trading activity in the futures market, declines with option volume, and is strongly predicted by the 30-day realized variance of the SC crude oil futures contract. External shocks, including the Russia–Ukraine conflict and the Geopolitical Risk Index, significantly elevate CNOVX levels. During the COVID-19 pandemic, mortality risk intensifies the volatility-amplifying role of futures trading and strengthens the volatility-dampening effect of options, while confirmed case counts have weaker influence. We further document a pronounced asymmetric leverage effect: negative futures returns raise CNOVX more than positive returns of equal size. However, volatility feedback effects are negligible, as changes in implied volatility respond primarily to contemporaneous market conditions. Overall, CNOVX serves as a timely and informative benchmark for monitoring risk in China’s evolving crude oil derivatives market, with valuable implications for investors, hedgers, and policymakers.

This study employs a Triple-Difference (Triple-DID) model, utilizing balanced panel data at the district level from Beijing, Shanghai, Shenzhen, and Hangzhou between 2018 and 2024, to critically evaluate the effectiveness of the Multi-School Zoning Policy (MSZP) in suppressing the capitalization of educational resources into housing prices and promoting educational equity. The research explicitly accounts for spatial and institutional heterogeneity as well as household strategic behavior.The results indicate that: (1) MSZP significantly reduced the average housing price premium associated with elite school districts by 15.2%, with the strongest effect observed in Beijing and the weakest in Hangzhou; (2) The policy's effectiveness diminishes as the spatial concentration of high-quality educational resources increases, highlighting persistent structural inequalities; (3) In areas characterized by resource monopolization and strong institutional inertia, the policy's suppressive effect on educational capitalization and its gains in educational equity are both constrained.The findings suggest that MSZP alone cannot fully overcome the "spatial lock-in" effect of high-quality educational resources. Achieving lasting equity requires complementary deeper institutional innovations, such as robust cross-district teacher rotation, transparent resource allocation mechanisms, and adaptive zoning algorithms. This research offers quantitative evidence for optimizing policy and institutional tools in the pursuit of comprehensive urban education reform.