The information asymmetry in financing has long been a challenge for small and medium-sized enterprises (SMEs), making it difficult for banks to accurately assess their credit risks. Moreover, in the traditional supply chain finance model, core bottlenecks such as data silos and the ineffective transmission of credit persist, greatly constraining financing efficiency and transparency. This study focuses on the application of blockchain technology in supply chain finance, exploring how blockchain can break through data silos, enable multi-tier credit circulation, and thereby enhance the financing efficiency of SMEs while reducing financing costs. Specifically, through normative analysis and case comparison, it examines how blockchain facilitates multi-tier credit circulation via digital debt instruments, builds trusted data pools to lower financing costs, and ensures data privacy in collaborative sharing. The findings indicate that blockchain technology can effectively rebuild the trust mechanism in supply chain finance, but its widespread adoption still faces key challenges such as technical interoperability, insufficient momentum for ecosystem development, and the absence of a legal framework. In the future, blockchain will integrate deeply with the Internet of Things (IoT) and artificial intelligence (AI), driving supply chain finance toward an intelligent, automated digital financial ecosystem.

With the intensification of retail competition, Sam's Club, as a representative of membership-based retail, faces the problem of extensive customer segmentation and inefficient marketing. This study takes Sam's Club membership data as the research object, and focuses on the research theme of customer segmentation and precision marketing strategy formulation based on random forest. By adopting methods such as data preprocessing, random forest modeling, feature importance analysis and data visualization, it solves the core research problems, including how to realize multi-dimensional accurate customer segmentation and how to formulate targeted marketing strategies. The results show that the random forest model can effectively divide Sam's Club members into four types: high-value loyal customers, high-potential growth customers, general-value stable customers and low-value churning-risk customers. The formulated differentiated marketing strategies can provide practical reference for Sam's Club to optimize customer operation and improve marketing efficiency.

Amidst the intensifying US-China geoeconomic competition, Southeast Asian states face mounting external pressures that threaten their economic autonomy. This study investigates whether performance legitimacy can sustain a nation's economic autonomy under such conditions. Moving beyond treatments of Southeast Asia as a homogenous bloc, this paper develops a two-dimensional analytical framework by integrating Performance Legitimacy theory with the Inclusive Developmental State theory. This framework deconstructs performance legitimacy into two core dimensions: the source of performance and the transformation of performance. Based on these dimensions, a 2x2 typology is constructed, yielding four ideal types: Endogenous-Inclusive, Endogenous-Unbalanced, Dependent-Welfare, and Dependent-Fragile. Using empirical data from 2000 to 2023, the study conducts a comparative analysis of four ASEAN founding members: Singapore, Malaysia, Thailand, and the Philippines. The findings reveal that performance legitimacy's impact on economic autonomy is conditional. Only the Endogenous-Inclusive type, characterized by domestically driven growth and equitable welfare distribution, successfully maintains the highest level of economic autonomy through dual material and political buffers. Endogenous-Unbalanced and Dependent-Welfare achieve only moderate autonomy, while the Dependent-Fragile type exhibits the lowest autonomy due to the absence of both buffers. This research contributes theoretically by unpacking the black box of performance legitimacy and offers a nuanced comparative framework to explain the divergent strategic behaviours of Southeast Asian states in response to major-power competition.

The rapid development of the mobile internet has turned short-video and live-streaming platforms, represented by Douyin, into major consumption scenarios and traffic hubs, generating massive real-time transaction data in the process. Drawing on quantitative analytical frameworks commonly used in financial research, this paper takes Douyin e-commerce activity indicators as the core explanatory variable and selects three industries—apparel, catering, and agricultural products—as research samples. Using a fixed-effect panel regression model and multi-case quantitative comparison, this study examines the heterogeneous impacts of Douyin e-commerce on the real economy based on 501 firm-year observations covering 2023 to 2025. Empirical results show that Douyin exerts a significant substitution effect on standardized retail industries such as apparel. In contrast, it generates significant promotion and integration effects on experience-oriented industries and agricultural products with strong offline value. The key factors shaping the direction of these effects include the irreplaceability of offline experiences and market efficiency. This study integrates quantitative financial methods with digital economy research, offering a measurable analytical framework for the coordinated development of digital platforms and the real economy.

The global retail industry is being reshaped by margin pressure, omnichannel competition, and changing customer expectations. Within this environment, Costco is an important case because it combines a low-price warehouse model with strong customer retention and relatively stable profitability. This paper examines Costco's financial performance and strategic outlook through a case study and comparative analysis. The analysis draws on recent retail literature and on the 2022-2024 annual reports of Costco, Walmart, and Target. Three analytical dimensions are emphasized: the economics of the membership model, operational efficiency, and the strategic implications of digital transformation and international localization. The findings indicate that Costco's strength does not come from high merchandise margins. Instead, it is created by the interaction of low-margin pricing, rapid inventory turnover, high member renewal, and disciplined human-capital investment. Membership fees remain a major earnings stabilizer, while operational efficiency allows the company to compete aggressively on price. At the same time, important risks remain, especially market saturation in mature regions, slower digital development relative to major rivals, and the need for deeper localization in overseas markets. The essay concludes that Costco's long-term resilience will depend on preserving its membership value proposition while accelerating technology adoption and locally responsive expansion.

Accounting information quality is a core factor affecting investor decision-making and capital market efficiency. In recent years, frequent financial fraud cases have intensified concerns over how deteriorating accounting information affects investor behavior. This study examines the Luckin Coffee financial fraud case using the event study methodology to investigate whether poor accounting information quality triggers negative investor reactions and whether market responses differ between the fraud suspicion disclosure stage and the official confirmation stage. Based on stock trading data from IPO to pre-delisting, two key event dates and a [-5,+5] event window are selected. Abnormal returns (AR) and cumulative abnormal returns (CAR) are calculated using the market model. The results show significantly negative AR and CAR at both stages, with a substantially stronger reaction during the confirmation stage. The findings highlight the severe market consequences of accounting fraud and provide implications for firms, investors, and regulators in strengthening information disclosure and market supervision.

The health-oriented transformation of the fast-food industry faces a persistent dilemma: consumers widely associate fast food with artificial additives and preservatives, making it difficult for brands to rebuild trust through conventional marketing strategies. This study examines Burger King's "Moldy Whopper" campaign as a representative case of reverse marketing that challenges such stereotypes. Using a single case study approach, the research analyzes the campaign's strategic logic and the underlying consumer psychological mechanisms from the perspectives of cognitive conflict and value reconstruction. The findings suggest that by deliberately exposing an apparently negative product attribute, natural mold growth due to the removal of artificial preservatives, the campaign disrupted entrenched consumer perceptions and generated strong public attention. This "self-stigmatization" strategy triggered cognitive dissonance, prompting consumers to reassess the relationship between food quality and industrial additives. The study demonstrates that reverse marketing can function as an effective trust-rebuilding mechanism in the fast-food sector and offers strategic insights for brands seeking to communicate health-oriented transformation in highly competitive markets.

Financial market sentiment has a crucial impact on market volatility and investment decisions, but existing research are limited by sufficient adaptability to multimodal data and incomplete sentiment analysis frameworks, which restrict their applicability in real-world financial scenarios. This paper proposes a targeted financial market sentiment analysis system that addresses the characteristics of multimodal financial data through dedicated data collection and preprocessing methods. A feature-level fusion framework based on a cross-modal attention mechanism is designed, and its effectiveness is validated through comparative experiments. Furthermore, a multi-level sentiment modeling architecture consisting of a base layer, intermediate layer, and application layer is constructed to support real-time sentiment classification, short-term prediction, and anomaly detection. On the basis, the system architecture, engineering implementation, and interface design of a sentiment analysis engine are completed. Experimental results demonstrate that the proposed fusion framework and hierarchical modeling system achieve superior performance in financial scenarios. Overall, this study overcomes the limitations of traditional single-modal sentiment analysis and provides a practical and scalable technical path for investment decision-making and risk management in fintech application.

Currently, data trusts mainly develop along two paths: the "information trustee model" and "third-party data trusts", and compared with conventional trust relationships, the introduction of the data controller role differs. When exercising rights on behalf of data subjects, data trustees should neither abuse their power nor supervise the compliance of data controllers' behaviours; At the same time, they need to participate in strategic interactions around data use rights. Considering that different trust objects have specific economic benefits; At the same time, because it involves personal information privacy and other rights issues, the interest demands of the participants cannot be overlooked. Therefore, the purposes for setting up a trust no longer solely encompass the protection of assets; instead, they now include the safeguarding of privacy rights. Given this situation, the fiduciary obligation of the data trustee will cover the entire life cycle of the data. The fiduciary obligations of data trustees mainly include: the obligation to ensure the security of personal data, the obligation to ascertain that there is no illegality in the purpose of using personal data, the obligation to specify the scope of exercising rights over personal data, and the obligation to avoid conflicts of interest. In actual application, the fiduciary obligations of data trusts are constrained by public interest factors and limits on the disposition rights of data owners.

To achieve carbon peak and neutrality, provincial governments employ policy mixes including regulatory constraints, market incentives, and innovation guidance to drive corporate green transition. Through a multi-case comparison of Zhejiang Longsheng, Hikvision, and Wenergy Group, this study analyzes the transmission pathway of "policy mix - business model reconfiguration - green transition." Findings show that different policies drive firms to reconfigure various dimensions of their business models via differentiated mechanisms, leading to distinct transition outcomes. Regulatory constraint policies force operational process reconfiguration, achieving process optimization (e.g., Longsheng's annual carbon reduction of 11,400 tons). Innovation guidance policies drive value proposition expansion, achieving technology-enabled transition (e.g., Hikvision's carbon emission density decreased by 23% over three years). Strategic planning policies drive business portfolio adjustment, achieving strategic reconfiguration (e.g., Wenergy's new energy installed capacity increased 32-fold over four years). The study suggests that policy mixes need to match corporate endowments to effectively drive green transition.