The objective of this study is to examine the predictive relationship between the China stock market and the bond market. With the worsening economic environment entering 2024-2025, the stock-bond linkage mechanism has played an increasingly important role in financial forecasts. But through the current forecasts, it has been noticed that the forecasts often ignore the systematic connection of market information between the markets. The construction of the model was focused on a VAR model with the return of the stock market as an exogenous variable, which was contrasted with the traditional ARIMA model, and the nature of the connection between the two markets was evaluated through Granger causality test, variance decompositions, and impulse responses. The outcome of the test showed the superiority of the VAR model in terms of the key predictive accuracy of the forecast, especially regarding the declining trend of the bond market yields. Overall, the findings suggest that the stock market exhibits significant leading and explanatory power over the bond market, indicating that integrating cross-market information can enhance forecasting accuracy and provide more effective quantitative support for asset allocation and risk early warning by leveraging leading indicators from the stock market.

With the development of modern society, more and more young people like to add some tea drinks, desserts and other leisure foods to their fast-paced lives. However, it is precisely because of this that the tea beverage industry is becoming increasingly popular and there are more and more brands, which leads to serious homogenization. The similarity between various brands is high, but the uniqueness of the brands is limited, making it difficult to highlight the advantages of their own brands and attract consumers. The competitiveness of the tea beverage market is increasing, and there are also many shortcomings between companies and departments. It is necessary to improve the company's marketing strategy in the market according to the constantly changing market situation, which will help the company increase profits, occupy advantages in the market, and enhance its position in the market. This article focuses on the optimization analysis of marketing strategies, combined with the current market situation, and takes Mixue Ice City as the research object to conduct research. The literature collection method and SWOT analysis method are used to deeply analyze the internal and external situation of Mixue Ice City. Based on the P marketing strategy, corresponding marketing optimization methods are proposed. This article provides an optimization plan for the marketing strategy of Mixue Ice City, which can also serve as a reference for other tea beverage industries.

Social media has become a constituent of life for young people in the United States. At the same time, anxiety disorders are common among this group. This study examines how social media use is linked to anxiety symptoms in youth aged 13–25. Based on existing research, the paper identifies three main psychological mechanisms: Fear of Missing Out (FOMO), upward social comparison, and cyberbullying. These mechanisms are shaped by platform features such as endless scrolling, curated content, and public feedback. Together, they contribute to negative outcomes including social withdrawal, sleep disturbance, and reduced academic performance. The study further discusses how these effects interfere with daily functioning and increase anxiety levels. To address these problems, a multi-stakeholder intervention framework is proposed. This framework includes digital literacy education in schools, well-being-centered platform design, and stronger mental health support from families, schools, and professionals. By focusing on specific psychological pathways rather than general screen time, this study provides a clearer understanding of how routine social media behaviors may increase anxiety and offers practical directions for prevention and intervention.

This study is on the dualistic effect of social media information disclosure in the current consumer purchasing decision-making process. Focusing on the beauty and personal care industry, a sector heavily influenced by social media, the paper describes the current landscape of information disclosure, which includes user-called-out reviews, influencer endorsements, brand-generated content, and corporate sustainability communications. The analytical component develops into three main positive effects: enhanced trust through transparency, relational social proof, and improved consumer literacy. The analysis goes into three relevant issues: information overload, the proliferation of misinformation, and the reinforcement of algorithmic bias and filter bubbles. To tackle the challenges, the study proposed a strategic framework implementation of information curation systems that are user-centric, enhancement of multi-stakeholder verification and regulatory enforcement mechanisms, and promotion of algorithmic transparency. The conclusion restates some of the more critical points about the importance of a well-balanced approach towards the release of information to the health of the digital economy. The study has limitations, mainly the use of secondary data, so future research should be done empirically in the form of an experimental design and longitudinal research. This paper contributes to businesses, regulators, and consumers to drive a more trustworthy, efficient, and satisfactory digital marketplace.

Fluctuations in global food prices pose a significant threat to economic stability and food security. Accurately predicting benchmark indices, such as the Food and Agriculture Organization's Food Price Index, is particularly crucial for policymakers and market participants. This article applies the classic automatic regression composite moving average method to the complete monthly FFPI data from 1990 to 2025. This study aims to address everyone's demand for a stable and easy-to-understand predictive benchmark. According to the Box-Jenkins framework for analysis, first, the first-order difference is used to stabilize the data. Then, it is determined that the ARIMA (1,1,0) model is the optimal one. This model has passed the key diagnostic check, and the residual also conforms to the characteristics of white noise. The out-of-sample prediction results calculate that the average absolute percentage error is 17.05%. The recent prices have been relatively stable. The discussion section explains the economic impact of the discovered price persistence, places the prediction accuracy within the volatility of the commodity market itself, and emphasizes the value of this model as a transparent baseline for evaluating more complex methods. This paper establishes a concise ARIMA (1,1,0) model as a key benchmark for FFPI predictions. It can serve as a practical and easy-to-understand tool for stakeholders, and also act as a fundamental baseline for fairly evaluating more complex models in future research.

Disposal effect is a classic phenomenon in behavioral finance, which refers to the irrational behavior of investors who tend to sell profitable assets prematurely and hold losing assets for a long time. Based on the theory of behavioral finance, this paper examines the existence and formation mechanism of the disposition effect in China's capital market and draws the relationship between net value volatility and investor behavior by analyzing the relevant data of the Blue Chip Select Mixed Fund of Efunds Fund, the leader of China's public fund industry. By comparing the extent of the disposition effect in different markets, an in-depth study of the disposition effect in the fund market not only helps to understand the behavioral patterns of investors and fund managers, but also reveals the micro foundations of the behavior behind macro issues such as market volatility, fund performance persistence, and systemic risk. The paper concludes with relevant recommendations based on improving the impact of the disposition effect on the market.

Under the backdrop of the corss-border e-commerce becoming a new motivation of the foreign trade growth, in order to save the issue of insufficient supply chain flexibility, the bottlenecks of the operational efficiency, and digital of JD's com's cross-border e-commerce sector, this study conducts a systematic case analysis of JD .com's cross-border e-commerce supply chain was based on supply chain resilience and the development of bottlenecks. This research sorts out the supply chain resilience advantages of JD.com, the layout of the "bonded warehouse + oversea warehouse" and the disadvantages of insufficient warehouse coverage, inaccurate risk prediction and insufficient data sharing. The paper deeply analyze the core causes of the lagged supply chain efficiency response, weak risk resistance resilience, and the problem of compatibility between technology and mode. The study finds that the insufficient cooperation between technology and processes, the lack of ecological resource integration, and the lag in organizational design are the root causes of the problems. Beside, this study indicate the optimization mechanism of supply chain resilience for cross-border e-commerce, and provides practical references for JD.com and similar companies to improve supply chain resilience through AI integration, ecological cooperation, and organizational adaptation.

It is quite a challenge to predict the prices of Bitcoin, and this is due to inconsistent movements and disobedience to linear trends that characterises the markets. This study created and experimented with a multimodal model that incorporates both the technical market indicators and macroeconomic forces. In this paper, a diverse collection of 95 engineered features was compiled and applied to test various architectures, including traditional linear models, as well as XGBoost and LSTM networks. This research placed these models under intense rolling-window testing and divided them by market cycles to test their weight in relation to performance. The research has discovered an obvious synergy: combining the various sets of data enhances the price projections, which is much more effective than when isolated metrics are used. XGBoost, specifically, was unique due to its strength. It is more accurate than other systems and performs more stably during bear markets. All such signals are not just hypothetical. When applied to actual trading contexts, the framework produced better risk-adjusted returns, characterized by a significantly healthier Sharpe ratio and less pronounced drawdowns. This suggests that Bitcoin is not merely a speculative bubble, but rather a two-tiered process driven by both short-term market perceptions and long-term economic shifts. This will offer institutional and private investors a real advantage in negotiating the turbulence of the cryptocurrency sphere by providing both precision and logic in their approach to this problem.

With the increase in trade along the Belt and Road, the Asia-Europe Container Line and the China-Europe Shuttle have become important corridors connecting the industrial supply chains of Asia and Europe. The frequent unexpected incidents are constantly striking the resilience of the Asia-Europe supply chain. The choice of corridors for Asia-Europe supply chains has become a key factor in supply chain stability. This study, according to the theoretical framework of supply chain resilience, adopts the case study method and literature analysis method, and conducts a research on the synergistic mechanism between Asia-Europe container routes and China-Europe liner. The results of the study show that the Asia-Europe container route and the China-Europe liner face significant vulnerabilities in terms of reliability, redundancy, stability and recoverability under a single transport corridor structure. The substitution and complementary mechanism between the two channels can effectively mitigate the risks associated with the failure of a single channel and improve the resilience of the supply chain. The conclusions of the study show that the Asia-Europe supply chain should strengthen the synergistic mechanism of the sea-railway dual-channel. Specific strategies include functional cargo diversion, structural corridor backup and enhanced coupling between nodes to optimize the transport corridor layout. These approaches enhance redundancy, increase flexibility and strengthen reliability and stability to improve overall Asia-Europe supply chain resilience.

This study conducts a comparative evaluation of the Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing (ETS) models for forecasting China's monthly inflation rate, based on Consumer Price Index (CPI) data from December 2022 to November 2025. Within a univariate forecasting framework, both models are estimated and assessed using out-of-sample accuracy measures, including the Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Scaled Error (MASE). The empirical results consistently show that the ARIMA model, specifically the ARIMA (1,0,0) specification, outperforms the ETS model across all evaluation metrics. This advantage reflects ARIMA's effectiveness in capturing short-term persistence and lag dependence in the inflation series, rather than increased structural complexity. In contrast, the automatically selected ETS (A, N, N) model produces smoother forecast trajectories by emphasizing level-based smoothing, which provides a stable representation of underlying inflation behavior but reduces responsiveness to short-lived fluctuations. Overall, the findings highlight the importance of aligning model selection with data characteristics and forecasting objectives. While ARIMA models are better suited for short-term inflation monitoring under low-volatility conditions, ETS models may remain informative for medium- to long-term trend analysis. Limitations related to the short sample period and the univariate framework suggest avenues for future research using extended datasets and multivariate models.