Data elements are a new driving force for innovation in high-tech industries in the digital age, and market integration is an external environmental support for promoting innovation in high-tech industries. It is necessary to study the impact of data elements and market integration on innovation in high-tech industries within the same framework. On the basis of theoretical analysis, this paper conducts empirical research based on China’s provincial panel data, and comprehensively uses panel regression model, mesomeric effect model and panel threshold model to study the impact of data element development and market integration on high-tech industrial innovation. The results show that: 1) The development of data elements has a significant positive impact on innovation in high-tech industries, and market integration has strengthened the innovation driving effect of data elements. 2) Market integration has a significant mesomeric effect, and data elements can drive high-tech industrial innovation by promoting market integration. 3) The threshold effect indicates that as the threshold value of market integration increases, the impact of data elements on high-tech industry innovation shows a trend of first increasing and then decreasing. When the market integration is at a moderate level, it is more conducive to unleashing the innovation driving effect of data elements. 4) Multidimensional regression analysis shows that the positive impact of data element development on high-tech industry innovation in the central region is significantly greater than that in the eastern and western regions. Compared with medium-sized enterprises, data element development has a stronger driving effect on the innovation of large high-tech enterprises. Therefore, high-tech enterprises should continuously enhance their ability to mine, apply, and transform data elements. Relevant government departments should pay attention to the development and utilization, efficient circulation, and ownership protection of data elements, actively cultivate a unified and standardized data element market, promote regional collaborative development of data elements, and strengthen the innovation driving effect of data elements.

As an emerging economic community with distinctive competitive advantages, the platform ecosystem has garnered significant attention in academic research. However, due to differences in research perspectives and contexts, a clear and unified theoretical framework has yet to be established. Based on bibliometric analysis and a systematic literature review, this study examines the existing literature on platform ecosystems, clarifies the concept and characteristics of platform ecosystems, summarizes the theoretical framework, and explores potential future research directions. The study first identifies that platform ecosystems encompass core elements such as modular architecture, value propositions, and ecosystem governance, alongside characteristics such as modular complementarity, non-hierarchical control, multi-party interactions, and network effects. It then constructs a theoretical framework for platform ecosystems, specifically elaborating on the foundational roles of modular architecture and value propositions, the governance structure formed by open access, power distribution, and benefit-sharing mechanisms, and value co-creation driven by mechanisms, processes, and value capture. Finally, based on the theoretical framework, the study proposes future research directions, aiming to provide valuable insights for both theoretical research on platform ecosystems and practical management applications.

Under the “dual carbon” goal, designing a carbon price mechanism that can not only promote the development of the carbon market, but also drive people’s low-carbon life and improve the return on investment in green industries has become a research focus. Therefore, this paper conducts a dynamic evolution of carbon price through multi-agent gaming toward the “dual carbon” goal. Firstly, the impact pathways of multiple mechanisms-including carbon quota, covered sectors, carbon inclusion, and carbon sink-on carbon pricing are analyzed. Secondly, dynamic development pathways for these mechanisms are designed to cover the entire carbon control process. On this basis, a dynamic carbon pricing game model is constructed, incorporating multiple stakeholders such as government regulation, carbon allowance suppliers, and demanders. Finally, simulation analysis based on simulated data yields the following findings: 1) Carbon price overall exhibits a three-phase dynamic evolution characterized by “decline, then rise, peaking, and finally stabilizing”. 2) Driven by the scarcity of carbon emission rights, the government’s free carbon quota allocation coefficient shows a year-on-year decreasing trend during the 14th to 15th Five-Year Plan periods. 3) Carbon price responds asymmetrically to policy tools: relaxing free carbon quotas or narrowing sectoral coverage leads to a decrease in the equilibrium carbon price, while weakening carbon sink and carbon inclusion mechanisms drives the carbon price upward. This study provides theoretical insights for understanding the evolution of carbon pricing and constructing a phase-adaptive policy system for the carbon market.

The parallel implementation of carbon trading and green certificate trading policies is of great significance to achieving the “dual carbon” goals of the power industry. Considering the complex process of policy action, based on the extraction of key policy elements and theoretical analysis, multi-agent modeling and simulation are used to simulate the systemic impact of heterogeneous micro-agent decision-making linkage on the low-carbon transformation of the power industry under different scenarios. The study found that: 1) The parallel implementation of carbon trading and green certificate trading policies has an incentive-compatible effect, but there is a conflict in the realization of policy goals. Under the baseline scenario, the power supply structure goal can be achieved. 2) In the short term, the parallel implementation of policies will cause more profit losses to the power industry, but from the perspective of achieving the long-term goal of carbon neutrality, the unit emission reduction profit loss rate is lower than the implementation of a single policy. 3) Carbon quotas, carbon prices, penalties, and technology subsidies are all key factors affecting the effect of policy parallelization. In the process of policy optimization, the implementation of a benchmark carbon quota allocation mechanism, 80 yuan/ton carbon price control, mandatory green certificate trading, and clean technology subsidies can achieve the “dual carbon” goals with lower industry profit losses. It is recommended that when taking emission reduction measures in the power industry in the future, the coordinated design of different policy entities, policy objectives and policy contents should be strengthened, and the policy combination advantages of carbon trading and green certificate trading should be fully utilized.

Based on a computable general equilibrium model with a carbon trading module, this study depicts the distortion factors in the power market, thereby exploring the socio-economic and energy-environmental effects of different industry expansions in the carbon market under the context of power market reform. The results show that: While maintaining the status quo of the power market, ① only covering the power sector has the larger negative impact on national GDP and households’ welfare, and the profit increase of the eight industries is the smaller; however, with the industry expansion of the carbon market, the impact on the macroeconomy can be mitigated to a certain extent. ② Regarding the abatement costs, only covering the power sector has the highest marginal abatement cost (MAC) and sectoral abatement costs; as the carbon market industry expands, both costs will decrease; but the total abatement cost shows the opposite trend. ③ In terms of energy, when the carbon market is expanded to four or eight industries, it can promote the renewable energy share and electrification rate to a certain extent. Regarding the environment, the carbon abatement rates of the newly included sectors have increased significantly, and each scenario has a better synergistic emission reduction effect of pollutants. In the case of power market reform, ④ The relationship between advantages and disadvantages of each carbon market industry expansion scheme in social-economy, abatement costs and energy environment is basically consistent with that before the reform; In this case, MAC and total abatement costs will be increased.

For the selection of charging infrastructure construction modes, optimization of strategies, and allocation of profits for manufacturers of new energy electric vehicles in a duopoly market, this paper explores investment strategies for charging infrastructure and pricing strategies for new energy electric vehicles under various construction modes by developing a non-cooperative cooperative biform game model, which considers scenarios where two competing manufacturers choose either autonomous construction mode or cooperative construction mode. Furthermore, the analysis is conducted to compare the benefits of the manufacturers participating in charging infrastructure construction in the scenario where only one manufacturer invests in autonomous construction. The findings reveal several key insights. 1) Competition in pricing for new energy electric vehicles is driving manufacturers to expand their charging infrastructure and boost demands. Manufacturers using the autonomous construction mode are particularly sensitive to price competition due to higher construction costs. 2) As the investment coefficient for charging infrastructure construction costs enlarges, new energy electric vehicles pricing and infrastructure investment decrease. Only with low construction costs can manufacturers achieve a mutually beneficial charging network. 3) Effective competition in charging service fees between manufacturers can drive industry growth, but excessive fees may hinder manufacturer cooperation and consumer acceptance. This research offers valuable insights for manufacturers making decisions on charging infrastructure investments.

Urban transport carbon reduction is an important breakthrough to achieve carbon neutrality and climate change mitigation. This study takes the Guangzhou subway system as a case study to construct a comprehensive environmental-economic analysis model for public transportation using the long-range energy alternatives planning system (LEAP). The model simulates the evolution of carbon reduction potential and external economic effects in the metro system from 1997 to 2060. The results indicate that the emission reduction potential associated with the transformation of transportation modes surpasses that of energy technology transformation, and subway transportation can significantly reduce transportation carbon emissions in all scenarios. Notably, substituting private cars with subway transportation demonstrates the highest potential for carbon emission reduction, achieving a maximum reduction of 3811 kt CO₂ emissions at equivalent transport intensity — constituting 93.3% of the total carbon emissions from private transportation. Although the carbon reduction potential of subway is expected to decrease as electrification and vehicle efficiency improve, it is projected to remain significant until 2060. The carbon return for the urban subway system is anticipated as early as 2029, but the speed of infrastructure development and changes in transportation mode will impact this time point. The benefits derived from mitigating a portion of road transport emissions surpass the social costs incurred by the subway, resulting in a net annual profit of approximately 0.4 to 0.8 billion Yuan after 2030. By 2060, the cumulative benefit-cost ratio is projected to reach 2 to 7 times. Consequently, the paper puts forward three policy suggestions on promoting carbon emission reduction in China’s urban subway system.

Under the climate change context, carbon risks arising from carbon policy uncertainty and carbon price fluctuations have permeated social security fund investment portfolios. This study investigates the relationship between the active management level of China’s social security funds and their carbon risk exposure based on risk contagion theory, analyzing whether these funds are reducing carbon risk exposure by divesting high-carbon-risk stocks through asset allocation. The findings reveal that social security funds are actively managing their equity portfolios to reduce carbon footprints. The active management level influences carbon footprints through the clustering degree of market industry development rather than industry concentration levels. Higher market clustering correlates with enhanced carbon reduction effects from active fund management. Regarding carbon market conditions, active management demonstrates higher sensitivity to carbon price fluctuations compared to carbon policy uncertainty. This research not only provides recommendations for scientific allocation of social security funds but also offers decision-making support for achieving China’s dual-carbon goals (carbon peaking and carbon neutrality) through pension fund management. The results contribute to understanding institutional investors’ role in climate risk mitigation and propose an industry clustering-based analytical framework for assessing active management effectiveness in carbon footprint reduction.

Energy, foreign exchange and gold, as important strategic reserves, show high sensitivity to geopolitical risks. Using time-frequency network analysis based on quantile vector autoregression, this paper reveals the risk spillovers between energy, foreign exchange, and gold markets in the Middle East, South Korea, and Western regions. The effects of geopolitical risks on risk spillovers between these markets are also examined. The results show that: 1) There are significant risk spillovers between energy, foreign exchange, and gold markets, with notable differences across various time and frequency domains. 2) Geopolitical events significantly affect the pattern of risk spillovers in Korea, Europe and the U.S. regions, and the responses of these to geopolitical events show significant differences. 3) The risk spillovers between energy, foreign exchange, and gold markets exhibit distinct time-varying characteristics, and geopolitical risk not only exacerbates the overall level of risk spillovers, but also affects risk spillovers between different regions and markets. These findings hold significant implications for policymakers and investors in understanding the risk spillovers between energy, foreign exchange, and gold markets under the impact of geopolitical events, enabling timely strategy adjustments.

With the rapid development and widespread application of information technology, textual data such as financial news, investor comments, and social media sentiment are widely used in international oil price forecasting. However, existing studies often neglect the impact of analysts’ commentary sentiment on crude oil prices. To this end, this paper collects analysts’ comments on the international crude oil market from 2009 to 2023, constructs analyst sentiment indicator, designs a framework for forecasting crude oil returns with heterogeneous data from multiple sources, and employs machine learning algorithms to investigate whether analysts’ sentiment can improve the performance of international crude oil return forecasting. The empirical results show that analyst commentary texts contain important crude oil market prediction information, and the prediction performance of various models on crude oil return is significantly improved by adding analysts’ sentiment indicator. In particular, deep learning algorithms show better prediction performance, with the LSTM model having the most prominent prediction effect. In addition, the content of analysts’ comment contents contains more prediction information about crude oil market than the comment titles, and the total sentiment of comment texts better reflects the amount of information about the crude oil market prediction than the sentiment extremes, means and divergences.

The debate on whether there is an idiosyncratic volatility anomaly (IVA) in the stock market has received widespread attention and remains unresolved. This paper distinguishes between daily idiosyncratic volatility and monthly idiosyncratic volatility to retest the existence of IVA to explain this controversy. Based on the sample of Chinese A-share listed companies, we find that the debate on the existence of IVA stems from the inconsistent sampling frequency of empirical data, the daily IVA is significant, while the monthly IVA is insignificant, the reason for this disagreement is the information attenuation of historical samples. The daily IVA is largely explained by the return reversal of a subset of overpriced stocks with high idiosyncratic volatility, and the equity overpricing factor plays an important moderating role in daily IVA. The heterogeneity test results show that the daily IVA is more significant in stock portfolios with lower institutional investor shareholding ratio, larger retail investors share, lower short selling and analyst attention. The key reason for stock overpricing and daily IVA is the unbalanced transaction size of margin trading caused by strict short-sale constraints in Chinese capital market. The findings above can help people understand the academic debate about the existence of IVA, and also deepen investors’ understanding of the pricing anomalies in China.

Forward-looking information reflecting future performance and business forecasts provides significant reference for investors’ decision-making. However, due to the dispersion of equity, minority shareholders often lack effective means to participate in the governance of corporate information disclosure. The establishment of investor interactive platforms has enhanced the voice of minority shareholders, enabling them to exercise activism through proactive expression under regulatory environments. This study examines how the online voices of minority shareholders impact corporate forward-looking information disclosure and finds that online voices improve the clarity of forward-looking information, as evidenced by increased quantitative and decreased qualitative disclosures. Further analysis reveals that this effect is realized primarily through mechanisms of capital market pressure created by aggregated investor opinions and the regulatory attention imposed by regulatory authorities. Moreover, the effect is more pronounced when the topics of minority shareholders’ voices focus on business strategies, when the proportion of negative tone is high, and when the corporate response rate is high. This paper identifies the transformation of forward-looking information disclosure triggered by minority shareholders’ online voices under the context of investor interactive platforms, providing important practical insights for companies to understand and respond to the interest demands of minority shareholders, as well as for regulatory authorities to further strengthen investor protection.

Operational risk is one of the vital factors that affect the survival resilience and sustained development of enterprises. This paper builds on an interdisciplinary perspective of finance and geographics and uses the combined panel data of Chinese A-share listed companies and financial permit information to empirically test the effect of bank branch geographic expansion on corporate operational risk. The results show that branch geographic expansion can mitigate bank-company information asymmetry by reducing bank-company distance and intensifying bank competition, thereby mitigating corporate financial constraints, enhancing bank debt governance, and consequently reducing corporate operational risks. A residual decomposition analysis is carried out based on different risk-taking incentives of enterprises and finds that bank branch geographic expansion mainly reduces operational risks driven by speculation incentives and passive risk-taking but promotes risk-taking driven by innovation incentives. Further analysis finds that, bank branch geographic expansion exerts a stronger impact on firms that are newly established or belong to hi-tech industries. Bank branch geographic expansion and loose monetary policy works a synergistic effect in terms of reducing corporate operational risks. The effect of bank branch geographic expansion turns out to be stronger among firms that are less disciplined by capital market or located in regions with less-developed infrastructure or digital finance. Our findings provide important policy implications from the perspective of physical branch distribution of commercial banks as regards reducing corporate operational risks, ameliorating business environment and building economic resilience.

The question of how managers balance risk control and opportunity exploitation in financial decision-making for firm growth has yet been fully addressed. Drawing from ecological rationality, through a scenario experiment with real estate firms, we have three key findings. First, managers’ “risk-reward” heuristics align with rational risk control principles, necessitating a nuanced balance between internal and external financing amidst dynamic environments. Second, when encountering market opportunities, managers moderately amplify their risk-taking inclinations. Third, within competitive landscapes, managers are urged to adeptly navigate the challenges of acquiring crucial resources, considering their scope comprehensively and gauging acquisition probabilities accurately. This research enriches the understanding of the “heuristic cognition and environmental fit” principle within the realms of corporate financing and firm growth research.

Consumers inevitably face waiting times when using ride-hailing services, and prolonged waiting often leads to dissatisfaction and abandonment. Based on the mental accounting theory, this study explores how coupons, as an economic incentive, can motivate consumers to wait, with a focus on the differential impact between the two stages of “waiting for matching” and “waiting for pickup”. Using a fixed-effects regression model and215,334 order records from a leading ride-hailing platform in China, the results show that: 1) Coupons significantly encourage consumers to wait, with each coupon (with an average face value of 3.41 CNY) increasing waiting time by 11.6% (approximately 41 seconds), and this effect is primarily observed in the relatively predictable pickup stage; 2) Higher-value coupons are more effective than lower-value ones in motivating consumers to wait during both the matching and pickup stages; 3) Compared to peak hours, coupons are more effective in motivating consumers to wait during off-peak periods, though they have a limited effect on motivating waiting during the matching stage; 4) Women, low-income individuals, and consumers without private vehicles show a more positive response to coupons, and coupons have a stronger effect on economy vehicle rides compared to higher-end vehicles. The findings provide empirical evidence to support the optimization of coupon design and distribution strategies for ride-hailing platforms.

Aiming at the problem of high costs and losses in the end distribution of fresh products e-commerce retail parcels, this paper proposes a collaborative distribution model for fresh products relying on the platform by considering the losses and costs of fresh products at the same time. Firstly, we design the operation mechanism process of collaborative distribution of fresh products; Secondly, we fully consider the costs of fresh products losses and customer delivery time demand, construct the fresh products losses function, and then put forward the vehicle path optimization model of multi-party collaborative distribution of fresh products, and design the algorithmic framework of integrated $k$-means clustering and ALNS for solving the problem; Then we use the Shapley value to solve the alliance costs-sharing scheme; Finally we verify the alliance costs-sharing scheme through numerical experiments. Finally, numerical experiments are conducted to verify the effectiveness of the collaborative distribution model for fresh products in reducing costs and losses, and to explore the factors affecting losses and costs. The study shows that compared with individual distribution, the costs saving rate of collaborative distribution of fresh produce considering customer time can reach 19.44%$\sim $41.78%, and can reduce the fresh products distribution losses by 18.01%$\sim $42.10%, and the costs saving rate of each enterprise in the alliance is concentrated in 20%$\sim $45%. In addition, customer distribution and distribution models are the influencing factors of collaboration benefits, when enterprises pay more attention to the costs should be used in large models of vehicle distribution, when enterprises pay more attention to the quality of fresh products should be used in small models of vehicle distribution, and the customer is uniformly distributed, the collaborative distribution model is more effective.

This paper examines the live-streaming e-commerce supply chain composed of brand owners, live-streaming platforms and consumers, and investigates the interactive impact of platforms’ quality information disclosure and channel selection. Four game models are developed to reflect different selling models and information disclosure strategies: voluntary disclosure in reselling (RV), mandatory disclosure in reselling (RM), voluntary disclosure in agency selling (AV), and mandatory disclosure in agency selling (AM). The study reveals the following insights: product quality and user loyalty affect the optimal decisions and profits of all participants, while the selling model impacts how the live-streaming platforms acquire and disclose information. Mandatory information disclosure, compared to voluntary disclosure, significantly decreases streamers’ motivation to select products and hurt brand owners profits and consumer surplus. Brand owners need to consider commission rates and product attributes to choose the right sales channel, potentially achieving a win-win outcome with the platform under certain conditions.

In the information age, expert service firms often use multi-channel strategies and offer differentiated services. However, the nuanced interactions between user information asymmetry and congested negative externality challenge the efficiency of tiered service. This paper examines how firms can leverage data and informational advantages to optimize information mechanisms and pricing, affecting consumer beliefs, revenue, and social welfare. The study finds a non-monotonic relationship between optimal information strategies and offline service costs, suggesting that a single-channel strategy may be optimal. Firms can strategically obfuscate low-value consumers’beliefs to enhance perceived service value while optimizing information for high-value consumers to reduce online cannibalization of the offline channel. This hierarchical information push mechanism maximizes social welfare more effectively than traditional approaches. The findings offer guidance for multi-channel firms and policy recommendations for regulating information push mechanisms, boosting the omnichannel service growth in the information age.

A manufacturer providing trade-ins faces the decision of whether to refurbish old products or not and whether to resell them with new products in the primary market (single-market strategy) or resell them to a separate secondary market (dual-market strategy). In this context, we analytically examine two refurbishing-reselling decision models and obtain the corresponding refurbishing-reselling conditions and equilibrium outcomes. The results show that regardless of which market strategy the manufacturer chooses, with different manufacturing costs and customers’ willingness to pay for refurbished products, there exist four outcomes: No trade-ins, no refurbishing, partial refurbishing, and full refurbishing. Although adopting the dual-market strategy is an effective way to avoid the cannibalization effect between new and refurbished products and enhance the demand of new products and the consumer surplus of the new segment in the primary market, it is not always better than the single-market strategy. If the size of the secondary market is large, the manufacturer prefers the dual-market strategy; otherwise, the manufacturer prefers the single-market strategy despite the risk of demand cannibalization.

This paper addresses the issue of emergency supplies stockpiling considering the phased demand and dynamic distribution network when unexpected events occur. A two-stage stochastic programming model based on option contract framework is developed. To control decision-making risks, the model incorporates joint chance constraints to limit the probability of risks caused by decisions. The optimization objective introduces conditional value at risk to effectively balance the conflicting goals between emergency objectives and cost objectives. The model involves two-stage decision problems: 1) Strategic decisions in advance, including the determination of emergency supplies reserves and supplier selection; 2) Tactical decisions afterward, encompassing the allocation of emergency supplies. The model aligns well with the current context of frequent unexpected events worldwide, demonstrating significant practical value. To tackle this complex mixed-integer programming problem, this study employs the framework of Benders decomposition algorithm and incorporates the following improvement strategies:1) Demonstrating superior initial lower bounds; 2) proving enhanced optimality cuts; 3) adopting the branch-and-check strategy. Through large-scale random experiments, this paper verifies the advantages of the improved Benders decomposition algorithm in terms of both accuracy and efficiency compared to general solvers and the original Benders decomposition algorithm. Finally, through real cases of floods and landslides, this paper verifies the effectiveness and superiority of the dynamic option model in solving practical problems. By analyzing the risk parameters of the model, it is discovered that setting appropriate combinations of risk parameter values can optimize the benefits of emergency rescue operations.

In order to promote the high-quality development of China’s economy, the Chinese government vigorously promotes the deep integration and innovative development of the logistics and manufacturing industries(DIIDLMI). However, there is still a lack of systematic evaluation on the coupling level of DIIDLMI, and there is an urgent need for targeted evaluation research. Based on the connotation and characteristics of DIIDLMI, this paper establishes an evaluation index system for the coupling level of DIIDLMI. An evaluation method for the coupling level of DIIDLMI is designed by combining the entropy method and the coupling coordination degree model. This method effectively combines the advantages of high reliability and validity of the entropy method, and can measure the degree of coordinated development between the logistics and manufacturing subsystems. In addition, this paper uses kernel density estimation to demonstrate the distribution and evolution of the coupling level of DIIDLMI. This paper finds that firstly, the coupling level of DIIDLMI in economically developed regions is generally higher, and there is a clear spatial clustering trend. Secondly, from different dimensions of DIIDLMI, some central and western provinces with weak economic development have better coupling in process optimization and organization synergy, which may be related to local market concentration, contract environment, and enterprise network characteristics. This indicates that each region may have its own advantages, thereby DIIDLMI should be based on the differentiated resources and conditions of each region. Finally, this paper proposes policy recommendations from three perspectives, namely, industrial cluster construction, regional harmonious development, and regional differentiated development, to further promote DIIDLMI.

The fair judging of large-scale innovative competition entries is a complex systemic issue, and the evaluation results may deviate significantly from reality due to the scoring characteristics, subjective preferences and systematic errors of the judging experts. The fact that the works evaluated by each expert in the judging of large-scale competitions are different leads to the premise assumption for the application of the standardized scoring method — that the levels of works judged by each expert have the same distribution — may not be valid. In order to solve the above problems and optimize the judging scheme, the standard scoring method has been improved. Firstly, a class of cross-assignment problems is proposed. Secondly, the mathematical model of the cross-assignment problem is applied to distribute the contest entries so that any two experts have as many cross-reviewed entries as possible in order to construct a reliable pairwise comparison matrix. Then, a mathematical model for determining the experts’ weights is established based on the pairwise comparison matrix and approximation of the ideal points method, and the calculation formula of the improved standard score method is proposed based on the experts’ weights. Finally, the ranking difference degree, Spearman’s rank correlation coefficient, scoring error degree and the works’ controversy degree of traditional method, standard score method, entropy weight method, method in literature and the proposed method in this paper are analyzed through an experiment, and the results show that our proposed method is more scientific and reasonable compared to other methods.

With the explosive growth of Internet information and the advancement of knowledge graph technology, the knowledge scale has expanded significantly, and confidence scores can characterize the uncertainty of knowledge. However, most current knowledge reasoning methods for knowledge graphs ignore the uncertainty of knowledge and do not adequately consider the semantic information and interactions between different semantics. To address these issues, this paper proposes a semantic-aware uncertain knowledge reasoning model (SAUR), which captures semantic information in entities and facts of the knowledge graph to enrich the reasoning process. Firstly, exogenous semantic information is introduced to enhance entity representation learning through the interaction of local spatial semantics. Secondly, entity representation is used as prior information to further mine contextual information of knowledge through triple sequences. Finally, the SAUR model performs confidence learning on the uncertain information of the knowledge graph to reason about uncertain knowledge. Experiments on the public data sets CN15K, NL27K, and ppi5K demonstrate that the proposed method outperforms all baseline methods in confidence prediction and link prediction tasks.

Against the backdrop of the reform in scientific and technological evaluation under the guiding principle of “breaking the five unicorns and establishing new standards”, peer review plays a pivotal role. Especially when confronted with large-scale evaluation tasks, the strategy of grouped evaluation has gradually emerged as the core approach of peer review. However, due to the varying scoring scales among experts, issues such as ensuring fairness within groups, achieving equitable comparisons among groups, and accurately assessing the credibility of experts have gradually emerged as challenges. To address these challenges, this paper systematically investigates strategies for aggregating expert panel scores using data standardization processing methods, comparing and analyzing the maximum-minimum value algorithm, centralization algorithm, Z-score standardization algorithm, and the newly proposed mean normalization algorithm. Through mathematical statistics and empirical analysis, this paper delves into the effectiveness and underlying issues of existing data standardization strategies in the context of both intra-group and inter-group evaluation in scientific and technological evaluation. The research findings reveal that different algorithms exhibit distinct advantages in eliminating differences in expert scoring scales, enhancing evaluation fairness, and improving accuracy. This provides specific and targeted guidance for optimizing algorithms and methods. The paper offers theoretical support for the rational selection and application of data standardization algorithms, further promoting the continuous optimization and improvement of the scientific and technological evaluation system.

In the face of complex cybersecurity issues brought about by digital transformation and information technology development, many enterprises have begun to explore diversified security defense strategies, aiming to achieve comprehensive cybersecurity protection. This paper takes a multi-level perspective of inter- departmental and inter-enterprise connections and constructs a multi-related enterprise multi-level cybersecurity investment optimization model under the condition of fuzzy cyber risk-related information. Building upon this foundation, the study utilizes sensitivity analysis and rotational update algorithms to perform deterministic transformations on models containing multiple fuzzy parameters and obtain optimal equilibrium solutions. The research indicates that sensitivity analysis effectively identifies critical fuzzy parameters in the model, specifically the losses incurred by this enterprise due to transmission risks from other enterprise departments. Appropriately narrowing their fuzzy intervals can significantly enhance the overall satisfaction of the model, providing valuable insights for enterprises in formulating cybersecurity investment strategies under fuzzy information. Furthermore, enterprises can effectively curb the deterioration of cyber risk environment by formulating appropriate cybersecurity objectives, reduce excessive profit-seeking behaviors among enterprises, and help the market maintain a stable equilibrium state at an acceptable level, therefore, enterprises should reasonably plan their wealth utility objectives to promote the common benefits of participating subjects in the multilevel cyber. Finally, the parameterized fuzzy information scenario is compared with its deterministic scenario to verify the validity and stability of the model.

In the context of accelerating the global value chains restructuring, China, Japan, Republic of Korea, Australia, New Zealand and ten ASEAN countries signed the Regional Comprehensive Economic Partnership(RCEP), the largest in the world. How will RCEP reconstruct the global value chain, especially what opportunities will it provide for the development of China’s value chain? Firstly, this paper adopts the global multi-regional input-output model to analyze the evolution characteristics of GVC division in RCEP members and three regions (Asia, North America and Europe) from 1995 to 2018 from two dimensions of GVC participation depth and breadth. Then the structural model is used to measure the restructuring effects of RCEP tariff reduction on the depth and breadth of GVC participation. The results show that all RCEP members will benefit from tariff reductions, and most smaller economies will benefit more than larger ones. RCEP will alleviate the current recession of globalization and the declining trend of GVC participation in ASEAN, China, Republic of Korea and other RCEP members. Meanwhile, RCEP will significantly enhance the GVC forward and backward linkages between members such as China, Japan, Republic of Korea, ASEAN with the Asian region, and make these economies participate in GVC in a more regional way, thus promoting the restructuring of GVCs to a regional direction, especially the deepening development of Asian regional value chains. Our research provides important practical significance for China to accelerate the construction of a new development pattern of domestic and international double cycles with the help of RCEP to cope with the increasing uncertainty of the world economy.

Key strategic resources play a pivotal role in the game of great powers. To reveal the temporal and spatial patterns of the distribution of key strategic resources in China and summarize their evolutionary characteristics, this paper defines the fundamental concepts of key strategic resources, employs various research methods such as structural breakpoint tests and standard deviation ellipse diagrams, analyzing the temporal evolution stages and spatial evolution patterns of key strategic resource quantity and price from both temporal and spatial perspectives. Data analysis results indicate: The temporal evolution of key strategic resource quantity and price can be broadly divided into three stages, high price stage, low price stage, and rebounding price stage, with each resource exhibiting specific characteristics in stages influenced by various interacting factors. The temporal evolution of key strategic resource quantity and price manifests complexity, classifiable into four types: Quantity-and-price-driven, demand-driven, price-driven, and quantity-and-price-regulated. The distribution of supply and demand for key strategic resources in China follows a pattern consistent with production and living activities, presenting a “Low West, High East” trend centered around the “Beijing-Shandong” region. The spatial evolution of key strategic resource quantity and price demonstrates agglomeration characteristics, distinguishable into diversified or concentrated imports and diversified or concentrated prices. The research results provide key directions and specific data analysis guidance for the prediction modeling and decision analysis of various key strategic resources in China. The conclusions of this study are helpful for achieving high-precision quantity and price forecasting, macroeconomic regulation, risk warning, and emergency system construction for key strategic resources.

In recent years, global inflation has remained high. However, data indicates that the transmission of global inflation to China is insufficient. This paper attempts to explain this phenomenon from the perspective of import prices. The study estimates a mixed time-varying parameter vector autoregressive model and identifies the determinants of import prices. Variance decomposition reveals that before 2020, exchange rate shocks and import price shocks accounted for over 60% of import price fluctuations. The time-varying impulse response indicates that shocks in import prices lead to an immediate drop in both inflation and output, behaving more like a negative demand shock than a supply shock. Additionally, since 2010, the effect has diminished considerably, potentially due to factors such as reduced trade openness, the expansion of the service sector, and a growing preference for goods produced domestically. Counterfactual analysis suggests that the negative impact of import price shocks on domestic inflation primarily stems from the recession effects triggered by the rise in import prices. The fundamental conclusions of the paper hold under larger-scale time-varying parameter models, sign restrictions, subcategory import price indices, and alternative output and inflation indices. The paper provides important insights for central banks in addressing import price inflation.

In the process of rural revitalization of China, agricultural land transfer has become the basis of rural land concentration and agricultural industrialization. Thus, the corresponding issue of farmland transfer subsidies has been highly concerned by China. This paper considers two different subsidy decision-making modes: The central government implements subsidy decision-making, and the local government implements subsidy decision-making. In each mode, there are three different subsidy strategies: Subsidies to farmers, subsidies to enterprises, and zero subsidy strategies. This study finds that no matter whether the central government or local government implements subsidy decisions, there is an equilibrium of defrauding subsidies under the subsidy strategy for farmers. In other words, farmers will transfer land at a negative price to obtain subsidies. But this equilibrium will not occur on the subsidy strategy for enterprises. Moreover, when the equilibrium of defrauding subsidies exists, the subsidy decision implemented by the central government can still improve the social welfare, but the subsidy decision implemented by the local government may improve or reduce the social welfare. This study also finds that the income equivalence theorem exists between the subsidy strategy for farmers and the subsidy strategy for enterprises. Specifically, no matter who implements the subsidy decision, the two strategies are the same in terms of total subsidy amount, farmers’ income, enterprise profits, consumer surplus and social welfare. Farmers and enterprises prefer the local government to implement the subsidy decision, but the social welfare is higher when the central government implements the subsidy policy. In addition, this paper extends the model to the scenario of asymmetric information, and finds that the introduction of asymmetric information leads to the equilibrium of defrauding subsidies within the subsidy strategy for enterprises.

The rural clean energy multi-energy complementary system is a crucial physical infrastructure for promoting rural energy transformation and achieving carbon neutrality. This study examines the characteristics of rural energy supply and demand, as well as the industrial structure composition. It develops a multi-stakeholder collaborative operation optimization and benefit distribution model for the rural clean energy multi-energy complementary system. Firstly, based on the operational characteristics of various energy entities, a collaborative operation mode for the rural clean energy multi-energy complementary system is established. Secondly, considering the carbon green certificate joint trading mechanism under current clean development policies, a robust optimization scheduling model for multi-energy complementary systems is proposed. Finally, a multi-agent interest coordination optimization strategy model is constructed based on the recognition concept. The findings indicate that energy complementarity and mutual assistance within the rural clean energy multi-energy complementary system can maximize on-site energy consumption while effectively enhancing the benefits of participating in the carbon green certificate joint trading mechanism. The interest-balancing strategy ensures the demands of multiple stakeholders and considers factors such as risks and costs, thereby increasing operability and feasibility.

Cooperating with technology companies is an important mean for regional banks to develop fintech. Using the manually collected data on the cooperation between each commercial bank and technology companies, this paper assembles a panel dataset of 334 regional banks from 2011 to 2019 as a sample to construct a PSM-DID model, and empirically examines the economic outcomes of the cooperative development of fintech on regional banks. We find that although the cooperation can promote the bank’s loan growth, it will also reduce the bank’s return on assets and increase the non-performing loan ratio. These findings hold with numerous robustness checks and different attempts to mitigate endogeneity concerns. Furthermore, the evidence on the comparison of the effects of cooperation on retail loans and corporate loans and the comparison between the outcome of business cooperation and that of technical cooperation, suggests that there appears principal-agent problem in business cooperation about retail lending. The negative effects of principal-agent problem arising from the cooperation are more pronounced for the technology company (acting as an agent) operating its own loan business or having cooperated with multiple banks simultaneously, and for the bank (acting as a principal) with weak fintech capability or low power in cooperation. This paper not only contributes to the relevant research on the principal-agent problem of financial intermediaries in a unique context of technology-driven financial innovations, but also provides an important reference for scientifically evaluating the economic consequences of commercial banks’ mode choice of developing fintech.

With the development of fintech, the use of financial livestreams for investor education has gained increasing attention in the asset management industry. Based on data from Alipay financial livestreams and sampled investor fund transactions, this paper examines the impact of financial livestreams on investors’ asset allocation behavior. The study finds that watching financial livestreams significantly enhances investors’ portfolio returns. As the number of livestreams watched increases, investors’ holdings of risky assets and portfolio diversity both rise significantly. The educational content of financial livestreams, along with the social presence induced by likes and comments, significantly boosts investors’ risky asset holdings and portfolio diversification. Heterogeneity analysis reveals that financial livestream has a stronger effect on male investors, younger investors, and those with higher risk tolerance, and is more pronounced when investor sentiment is low. This research provides empirical support for the role of fintech in improving investor behavior and enhancing the inherent stability of capital markets.

Urban green space construction is a key project for the Chinese government to promote ecological civilization. This paper uses panel data from listed companies in China from 2004 to 2019 to calculate total factor productivity (TFP). Using fixed-effects methods, the paper investigates the impact of urban green space coverage on corporate production efficiency. The results show that urban green space can effectively improve corporate production efficiency. Dynamic effect analysis reveals significant heterogeneity in this positive effect across cities with different degree of air pollution and stages with different degree of resident environment awareness. The paper innovatively investigates the mechanism from the labor perspective: Urban green space coverage primarily enhances labor productivity and optimizes labor structure and increases corporate R&D investment through compensation substitution. This study reveals the labor mechanism between urban environmental optimization and corporate economic benefits and provides solid theoretical support and empirical evidence for implementing the concept that “improving the ecological environment is development productivity”.

Investors’ effective response to information serves as a crucial indicator for assessing the development of capital markets. Drawing on an information feedback perspective, this paper explores how individual investors with varying wealth levels trade in response to corporate announcements in the market, utilizing investor trading data provided by a prominent brokerage. The research reveals that less wealthy investors (below 500thousand) primarily focus on dividend and earnings announcements, and do not trade before announcements. More wealthy investors (between 500 thousand and ten million) engage in pre-announcement trading, have the ability to discern positive announcements while cannot accurately interpret negative ones, and have a risk preference. Those with the highest wealth levels (above ten million) demonstrate an ability to accurately anticipate market trends and the transactions are stable. The heterogeneity in trading behavior further manifests in investment returns, showing a gradual reduction in portfolio losses with increasing wealth levels. The results come from different investors’different timing of obtaining information and differences in market judgment. This study deepens the understanding of individual investor trading strategies, providing a reference for the education of individual investors.

The process of economic globalization and financial integration has not only made the links between international financial markets increasingly close, but also led to more complex risk linkages among global financial markets especially under tail events. Therefore, it is essential to study the risk contagion relationship in global financial markets to ensure the healthy development of regional finance. In this paper, we propose a nonlinear tail risk measure — CoEVaR$^{\mathrm {SIM}}$ by combining the single-index expectile model and conditional expected value-at-risk (CoEVaR). Based on our risk measure, we construct a dynamic risk spillover network of global stock markets using 30 important global stock market indices from 2009 to 2023, measure the systematic risk of global stock markets and explore the dynamic characteristics of the tail risk spillover structure. The empirical results show that: 1) From the perspective of risk measurement, the systematic risk index constructed by our risk measure has time-varying characteristics, which is very sensitive to tail risk events, and significantly increases during the crisis period. 2) From the perspective of the evolution of risk network, the risk trends in the Americas, Europe, and Asia are relatively consistent during low-risk periods. While during high-risk periods, the stock markets of the Europe and North America regions spillover risks to other regions’ stock markets. 3) In terms of event analysis, the COVID-19 epidemic in 2020 exacerbated the inter-regional risk spillovers in the global stock market. The Russian-Ukrainian war in 2022 dramatically altered the structure of risk spillovers in the global stock market, with the Russian stock market being the only center of risk exporter. This study can inform macro policy makers, multinational financial investment institutions and the prevention of offshore risks.

In order to explore the impact of air pollution on residents’ happiness, this study empirically analyzes the effects of subjective air pollution perception and objective air pollution on residents’ happiness as well as the transmission mechanism based on the data from the 2021 Comprehensive Survey of Social Conditions in China (CSS2021) and the China Statistical Yearbook. The results of the study found that air pollution has a significant negative effect on residents’ happiness. Specifically, objective air pollution itself does not have a direct and significant effect on residents’ happiness, but the objective air pollution can reduce residents’ happiness level by affecting their subjective perception of air pollution. It was further found that the heterogeneity of residents’ individual characteristics also moderated the effect of air pollution perception on happiness, with men’s happiness more likely to be negatively affected by air pollution perception, and low-educated and low-income individuals more likely to have impaired happiness under the influence of air pollution perception. This study provides new ideas for the comprehensive management of air pollution to improve the overall happiness level of residents and to narrow the “happiness gap”.

This paper investigates the Stackelberg investment-reinsurance game between insurer and reinsurer under information asymmetry. The reinsurer and the insurer play the roles of leader and follower, respectively. At the beginning of the transaction, both parties have insider information related to insurance claims at a future moment. The insurer believes this information while the reinsurer does not. The correlation between the insurance market and the financial market is also considered. Under dynamic mean-variance criterion, a system of expanded HJB equations is solved to derive the time-consistent investment and reinsurance strategies and value functions of the two parties in the reinsurance contract, respectively. Finally, some numerical examples are provided to analyze the effects of important parameters on the equilibrium strategies of the two companies. The results show that the reinsurance strategy of the insurer moves down with the accuracy of the insider information decreasing. The reinsurance premium strategy is affected by the insider information owing to the consideration of Stackelberg game, even if the reinsurer does not believe in the insider information, and the strategy increases as the accuracy of the insider information decreases. The investment strategy of the reinsurer is also related to the insider information, which is due to the consideration of market correlation. In the case of positive market correlation, the strategy moves up with the accuracy of insider information increasing. In the case of negative market correlation, the opposite is true.

Large-scale research infrastructures play a crucial role in the national innovation system by supporting innovation entities in overcoming scientific challenges, advancing technological innovation, and fostering economic and social development. Despite their importance, the evolutionary dynamics and internal mechanisms of collaborative innovation involving facility managers, enterprises, universities, and other innovation institutes remain poorly understood. This lack of understanding hinders the full realisation of the innovation potential of these infrastructures. This paper aims to address this research gap in order to enhance the performance of large-scale research infrastructures through multi-institutional collaborative innovation, thereby promoting the high-quality development of the national innovation system. The paper constructs a three-party evolutionary game model from the perspective of absorptive capacity, involving facility managers, enterprises, and universities. Evolutionary stable strategies are integrated to analyse the collaborative innovation mechanisms among these three parties, relying on large-scale research infrastructures, while numerical simulations are employed to identify the main influencing factors. The study reveals that collaborative innovation in large-scale research infrastructures exhibits distinct stages and dynamic characteristics: From the initial “isolation” dilemma faced by enterprises, to the developmental stage of multi-institutional collaboration, and finally to the mature stage of independent research and development by enterprises. Absorptive capacity is identified as the key driver for advancing through these stages. An increase in absorptive capacity enables enterprises to transition from reliance on collaborative innovation with facility managers and universities to leading independent R&D efforts. Additionally, the study identifies six types of collaborative innovation mechanisms and observes dynamic transformations among them, influenced by factors such as government support, the risk of trade secret disclosure, conflicting objectives, and service costs. The study recommends targeted policy interventions at different development stages: Providing full-service incentives during the start-up phase, reducing cooperation conflicts during the development phase, and offering enterprise-led support at the maturity stage. It also advises enhancing enterprises’ absorptive capacity and intellectual capital through talent acquisition and joint training, while promoting flexible service models and contracts to mitigate trade secret risks and address conflicting objectives and service costs.