ISSN 2512-6660 (Online) ISSN 1478-9906 (Print)
CN 10-1192/N
CN 10-1192/N
Accurate daily tourism demand forecasting is a fundamental input for effective revenue management and resource allocation but is often hindered by the complex, multi-scale seasonal patterns inherent in high-frequency data. This study addresses this challenge by systematically evaluating advanced decomposition-based forecasting models. We compare the performance of the trigonometric box-cox ARMA trend seasonal (TBATS) model and a hybrid complete ensemble empirical mode decomposition with adaptive noise-neural network (CEEMDAN-NN) model against traditional benchmarks. The findings reveal that both models consistently outperform conventional approaches. Notably, the CEEMDAN-NN hybrid model achieves the highest forecast accuracy, while TBATS offers strong interpretability for complex seasonalities. These results underscore the benefit of integrating decomposition techniques to improve the precision of daily forecasts, thereby supporting data-driven revenue management decisions.
Using a sample of Chinese listed companies from 2011 to 2023, this study empirically investigates the impact of how investor interactive supervision affects firms’ capacity utilization and the mechanism through which this effect occurs, addressing the gap in existing literature regarding how investor participation influences firms’ capacity decision in the presence of information asymmetry in capital markets. Methodologically, we construct an objective and fine-grained IIS indicator by leveraging 5.6 million posts from two official Chinese investor interactive platforms: We first expand supervision-related keywords via the word2vec algorithm before validating and classifying supervisory related posts using the Qwen2.5-72B large language model (LLM). Key findings reveal that IIS significantly enhances CU, with heterogeneous effects across firms. Mechanism tests confirms that IIS improves CU through two channels: Mitigating managerial myopia and alleviating financing constraints. This study contributes to the literature by extending research on investor interaction from financial performance governance to operational efficiency, pioneering the application of LLMs in corporate governance to measure supervisory behavior at a micro level and enriching information asymmetry theory by linking external investor supervision to firms’ capacity utilization.
Since China put forward the carbon peaking and carbon neutrality goals, the green and low-carbon transformation of the economy has attracted the attention of the whole society, and green bonds have developed rapidly under this opportunity. In the perspective of sustainable investment and ESG investment, green bonds have attracted more attention of investors, which have a crowding out effect with credit bonds in other industries. However, they are also affected by a series of common macro factors and thus change in the same direction. It is of great theoretical and practical significance to explore the spillover effect between different types of bonds including green bonds, and reveal their dynamic change characteristics. Based on the TVP-VAR model and DY spillover index, this paper constructs a correlation system including green bonds, climate bonds and other nine industry credit bonds, and discusses the evolution of systemic risk and green bond spillover effect from a dynamic perspective. It finds that systemic risk is greatly affected by external shocks, while the spillover effect of green bonds is heterogeneous in different periods and for different industries. The spillover effect of green bonds will be significantly strengthened in the period after the promulgation of green finance policies.
Driven by the urgency of technological innovation in China, science and technology (S&T) innovation platform supply chains have become increasingly important. This paper examines optimal government subsidy allocation and pricing strategies for such supply chains under two operating models: A public benefit model that maximizes social welfare and a market oriented model that maximizes platform profit. We develop a tripartite game-theoretic model involving resource providers, research users, and the S&T platform to determine optimal subsidy schemes and their effects on service charges, commissions, social welfare, and platform profitability. The results show that, in both models, the logic of pricing adjustment in response to changes in the innovation service cost coefficient remains unchanged whether subsidies are provided or not. Under the public benefit model, subsidies improve social welfare but reduce platform profits because the platform lowers user fees to satisfy public objectives. Under the market oriented model, subsidies to both resource providers and research users enhance social welfare and platform profits, with user subsidies generating a stronger marginal effect. In particular, subsidizing research users under the market oriented model is the optimal strategy, as it improves social welfare, maintains positive platform profits, and supports balanced pricing. Numerical analysis and sensitivity tests confirm these findings.
As a visual medium with immense dissemination potential, short videos have emerged as a highly promising tool in tourism marketing. Consequently, destination marketing organizations (DMOs) are increasingly adopting short videos to promote scenic attractions. Despite this growing practice, the core factors that drive user engagement with DMOs’ short videos remain underexplored, in particular the specific role of media richness. To address this gap, this study draws upon media richness theory and selects mount siguniang, a renowned Chinese scenic destination, as a case study. Employing analysis of variance (ANOVA), we systematically examine how video theme, format, and publisher attributes influence user engagement. Our empirical results indicate that factors including theme, duration, release timing, use of subtitles, background music, music genre, and publisher’s follower count significantly affect engagement levels. This suggests that richer media content correlates with higher engagement. Based on these findings, the study offers practical recommendations for enhancing tourism short video marketing strategies.
As a pivotal downstream product within the lithium resource industry chain, lithium batteries have increasingly captured the attention of international trade in recent years. Analyzing the current situation of international lithium battery trade, exploring the cooperation mechanism in the trade process and explaining the operating principle of trade cooperation are important ways to ensure trade stability and improve trade efficiency. Initially, this study employs complex network theory, utilizing global lithium battery trade data from 2015 to 2024 to construct a global trade network and analyze its distribution characteristics. Subsequently, based on the evolutionary game theory, the model incorporates price fluctuations in the lithium battery market and the heterogeneity of economic preferences among trading countries, integrating mechanisms for rewards and penalties, as well as variations in the trading environment, to develop an evolutionary game model for international lithium battery trade cooperation within a small-world network framework. Lastly, through numerical simulations, this analysis explores the internal factors of the trade system including initial proportions, cooperation benefits, cooperation costs, opportunistic behaviors such as “free-riding”, reputation loss, economic preferences and external factors, such as the complexity of trade networks, trade network scale, fluctuations in lithium battery market prices. The results reveal that from 2015 to 2024, the global lithium battery trade exhibited a positive development trajectory, with trade network patterns demonstrating increasing complexity. Enhanced cooperation benefits are illustrated to promote the diffusion of the evolution of system toward cooperation, though a specific threshold exists. Additionally, the introduction of a dynamic, multiple reward and penalty mechanism can further facilitate trade cooperation, and reputation loss factors can partially constrain opportunistic behaviors. Furthermore, while fluctuations in the lithium battery market price have the potential to induce trade instability, and the economic preferences of trading countries will further influence their attitudes towards cooperation in the face of price fluctuations. Ultimately, both complex trade networks and mature trade scales provide the optimal environment for lithium battery trade cooperation, exhibiting substantial risk resistance capacity. The study, by analyzing the dynamic game process of trade cooperation among lithium battery trading countries, reveals the long-term evolution law of the trade network, providing theoretical support and decision-making reference for the sustainable development of the lithium resource industry chain under the background of trade fluctuations.
Participation in agricultural cooperative organizations is an important form of industrial chain organization expansion, but also an important way to effectively reduce industrial chain transaction costs and promote the quality and efficiency improvement of new agricultural business entities. Based on the micro-survey data of 553 family farms in Xinjiang Corps in 2021, this paper uses stochastic frontier approach to measure the business performance of family farms in Xinjiang Corps, and discusses the action mechanism and difference in the influence of participation in agricultural cooperative organizations on the business performance of family farms. The research shows that participation in agricultural cooperative organizations can significantly improve the business performance of family farms through such mechanisms as easing the constraints of farm credit financing, reducing the transaction costs of agricultural products, and promoting the adoption of agricultural technologies. Compared with food farms, cotton and fruit and vegetable family farms have more obvious performance improvement in participating in agricultural cooperative organizations. Moreover, participation in agricultural cooperation organizations has a stronger effect on the performance improvement of farms in northern Xinjiang. It is found that the vertical extension of the industrial chain between “family farm + agricultural cooperative organization” and network marketers, chain supermarkets, industrial leading enterprises and agricultural production and marketing enterprises is conducive to the improvement of the business performance of family farms, among which the performance improvement effect of “family farm + agricultural cooperative organization + network marketers” mode is more obvious. The research provides a scientific reference for deepening the expansion of the industrial chain organization and promoting the quality-oriented growth of family farms in the new era.
