Table of Content

    20 December 2022, Volume 30 Issue 12 Previous Issue   
    An Empirical Study on the Systemic Risk of Chinese A-Share Listed Companies Based on Multi-layer Network
    ZHANG Fei-peng, XU Yi-xiong, ZOU Sheng-xuan, CHEN Yan
    2022, 30 (12):  13-25.  doi: 10.16381/j.cnki.issn1003-207x.2021.2666
    Abstract ( 312 )   PDF (7194KB) ( 517 )   Save
    Systemic risks are received much attention in financial studies these years. Most existing risk measures could be not directly appliable to measure the systemic risk contribution of financial institutes in China for the complex financial network in big data era. To describe the nonlinear correlation of financial returns, a new multi-layer correlation network, named by Local Gaussian Correlation Network (LGCNET), is constructed by combining quantile regression and local Gaussian correlation coefficient.The new method is used to measure the systemic risk contributions of 50 A-shares listed companies in China from 2018 to 2021. The empirical results show that: 1) The finance and technology industries are often the center of network nodes. They often have high correlations with other industry companies, which demonstrates that such industries are usually the center of risk transmission. 2) Due to their high market value, infrastructure and banking companies are generally more important in the financial system. Meanwhile, more attention should be paid to companies whose importance exceeds their market value, because they often have greater influence in the market. 3) At the systemic level, the whole system has a relatively high level of risk during 2018, especially at the beginning of 2018, which may be mainly affected by the increase of credit risk and the trade war, whereas the domestic systemic risks have been well controlled during the new crown epidemic in 2020. Finally, some suggestions are provided for improving China’s financial risk prevention system.
    References | Related Articles | Metrics
    Online Procurement Problem of Inventory-dependent Deteriorating Items
    ZHENG Meng, DAI Wen-qiang
    2022, 30 (12):  26-37.  doi: 10.16381/j.cnki.issn1003-207x.2021.2633
    Abstract ( 270 )   PDF (1660KB) ( 413 )   Save
    According to the common phenomenon that the larger the inventory is, the more attractive the purchasing interest of consumers is, a procurement model of deteriorating items is established whose demand rate is inventory-dependent. Under the condition that the future price obeys arbitrary distribution, an online procurement strategy of deteriorating items is proposed under price uncertainty. The online algorithm and its competitive analysis method are used for modeling and analysis, and an effective online competitive procurement strategy is designed. The theoretical competitive ratio and the optimal economic procurement quantity are compared with the optimal offline strategy. Finally, through the numerical analysis, it is proved that the strategy has an excellent competitive performance ratio in reality and performs well under multiple price series, which shows that the strategy is robust and can provide valuable and reasonable decision-making suggestions for firms.
    References | Related Articles | Metrics
    Distributed Multi-project Stochastic Scheduling with Two-stage Coordination Mechanism of Resources Allocation
    LI Fei-fei, XU Zhe
    2022, 30 (12):  38-51.  doi: 10.16381/j.cnki.issn1003-207x.2021.2641
    Abstract ( 260 )   PDF (2811KB) ( 528 )   Save
    Enterprises tend to manage multiple projects in a distributed manner where the global resources are shared among autonomous projects. In the actual process of multi-projects scheduling, however, the availability of global resources is often uncertain due to some unexpected situations or stochastic factors, resulting in disruption of activities, waste of resources and other consequences. A new level of complexity is therefore added to the traditional project management. The distributed resource-constrained multi-project scheduling problem (DRCMPSP) is studied under uncertain global resource availabilities.
    References | Related Articles | Metrics
    Research on the Evolutionary Game of Cooperative Innovation in the Core Firm-led Makerspace from the Perspective of Value Co-creation
    CAI You-hua, MENG Qiu-yu, CHEN Guo-hong,
    2022, 30 (12):  52-62.  doi: 10.16381/j.cnki.issn1003-207x.2021.2642
    Abstract ( 260 )   PDF (1864KB) ( 383 )   Save
    Makerspace has become a typical incubation platform to support the innovation and entrepreneurship development of start-ups. However, the rapid expansion of makerspaces has also led to frequent negative news, so a more effective management mechanism is urgently needed to promote the development of makerspaces. The core firm-led makerspace relies on its own strong industrial resource advantages and reasonable and effective operation mechanism, exudes development vitality, and is favored by many new start-ups. Taking the core firm-led makerspace as the research object, the cooperative innovation evolutionary game model between the operator and entry-settled enterprises is established. The influence of the changes of the profit and cost parameters of both parties on the evolutionary stability strategy is analyzed, and then the influence of the reward and punishment mechanism and the game results are analyzed with numerical simulation methods. The results show that the cooperative innovation between makerspace operators and settled firms is mainly affected by the total cost, the operator supervision cost, the cost sharing ratio, the additional revenue and the additional revenue distribution ratio. The result also suggests that margin, loss, penalty and incentive have positive effects on the cooperative innovation behavior, which indicates that the reward and punishment mechanism can effectively solve the dilemma of value co-creation. It is of great guiding significance to explore the influencing factors of collaborative innovation in core firm-led makerspace for further promoting the high-quality development of China’s makerspace.
    References | Related Articles | Metrics
    Forecasting Online Retail Sales of China Based on Splitting-filling-decomposition-ensemble Model
    ZENG Neng-min, , ZHANG Ming, YU Le-an, ,
    2022, 30 (12):  63-76.  doi: 10.16381/j.cnki.issn1003-207x.2021.2635
    Abstract ( 223 )   PDF (5233KB) ( 377 )   Save
    In recent years, China’s online retail industry has developed rapidly. Accurate prediction for online retail sales is the basis for government to formulate retail policies, as well as the foundation for ecommerce and logistics companies to determine operation strategies. However, no existing research has focused on macro online retail sales prediction driven by data characteristics. Forecasting the monthly online retail sales of China is a great challenge because the dataset has the characteristics of small sample size, high volatility, large holiday influence. In addition, the China’s online retail sales data has a unique data missing phenomenon: the total number of January and February is known but the monthly value is missing, which is caused by the relevant regulations of the National Bureau of Statistics. Motivated by these, a splittingfillingdecompositionintegration (SFDE) prediction framework is proposed. Specifically, firstly, the data set of online total retail sales of China is split into two parts, i.e., physical retail sales data and nonphysical retail sales data. Secondly, in the light of the incompleteness of online physical retail sales data, a revised spline interpolation approach (i.e., the hybrid approach of spline interpolation and dichotomous adjustment) is proposed to fill the missing value of the data. Meanwhile, considering that nonphysical retail data has different trends at different stages and increasing fluctuations, another revised spline interpolation approach (i.e., the hybrid approach of piecewise linear function fitting and spline interpolation) is proposed to fill the missing value of the data. Thirdly, based on the different characteristics between the physical retail data and nonphysical retail data, two hybrid ensemble forecasting approaches are proposed to predict the above two series, where the first one integrates multiplication decomposition, ARIMA and moving average, and the second one integrates STL decomposition, BP neural network and gray waveform forecasting. Finally, the prediction results of the above two series are integrated to get the predicted value of online total retail sales of China. In our experiments, the monthly data of China’s online retail sales from 2015 to 2019 are selected to verify the model performance. The results obtained in this study show that the revised spline interpolation approaches based on data characteristics are able to solve the problem of mentioned missingdata filling effectively. In addition, the combination of the revised spline interpolation approaches and the hybrid ensemble forecasting approaches achieved significant performance improvements over single model. Furthermore, the SFDE framework of combining the strengths of the conventional and deep learning methods provides a robust modelling framework capable of capturing the nonlinear nature of the complex online retail sales series and thus producing more accurate forecasts. On the whole, the proposed hybrid framework enriches the research of missing value filling methods, and have tremendous scope for application in a wide range of areas for achieving increased accuracies in complex time series forecasting.
    References | Related Articles | Metrics
    Research on Clustering and Assignment of Emergency Blood Donors for Emergency
    CHEN Xi, ZHANG Yi-fei, SUN Ya-ya, LIANG Hai-ming, ZHANG Wen-bo
    2022, 30 (12):  77-85.  doi: 10.16381/j.cnki.issn1003-207x.2019.1743
    Abstract ( 232 )   PDF (1282KB) ( 368 )   Save
    Various emergency events have happened frequently in recent years, which have caused severely impact on society and economy. Once the emergency events happen, it usually implies the need of emergency medical, especially the need of emergency blood. And the number of emergency blood donor will boom, inducing overloaded operations in blood collection agencies and facilities. An optimization method is proposed for emergency blood donor clustering and allocation, which can effectively solve the above problems. In the proposed method, the classification rules of emergency blood donors are introduced, following with an algorithm combining Canopy and K-means to cluster emergent blood donors. Furthermore, a multi-objective allocation optimization model is established accordingly in terms of blood donors, through which psychological factors areadopted, maximizing the demand for blood to meet and to balance the workload of various blood collection agencies and facilities, whereby the effectiveness of blood donors is optimized and maximized. A multi-objective optimization model is also established by taking into account the time preferences of emergency blood donors, thereby formulating a mechanism for blood donation. Throughout performing an optimization algorithm to solve the model, an emergency blood donor allocation profilecan be obtained. Consequently, related case analysis is conducted to validate and verify thefeasibility and effectiveness of the method.The obtained results in the example show that the proposed method can contribute to the classification and distribution of emergency blood donors, and further promote the satisfactions of the emergency blood donors.
    References | Related Articles | Metrics
    Analysis Model for Infrastructure Resilience Based on Linear Piecewise Recovery Function——A Case Study of C County Power Network
    GAO Lei, GONG Jing
    2022, 30 (12):  86-95.  doi: 10.16381/j.cnki.issn1003-207x.2019.1887
    Abstract ( 214 )   PDF (2238KB) ( 375 )   Save
    Recent disasters,such as earthquakes, typhoons, and terrorist attacks, disrupted operations of critical infrastructure systems and even destroyed functions of our society. Therefore, it is necessary and important to maintain critical infrastructure systems which are not only cost-effective but also able to respond quickly, handle smoothly, and recover promptly from disruptions. To achieve this goal, a linear piece-wise recovery function is proposed based on three other recovery functions and then develops a resilience analysis model. The proposed model is applied to the decision-making processes of improving the resilience of the power system in C county. C county is a coastal area vulnerable to hurricanes and its power network consists of 958 arcs and 939 nodes, including 4 supply nodes and 56 transfer nodes. In addition, to estimate the probability of nodes being disrupted and identify the critical nodes that require priority protection, different disaster scenarios are simulated according to the type, severity, and extent of the disaster in C county. The results show that: (1) this linear piece-wise recovery function enables tradeoff between a cost minimum system and a resilient system; (2) this resilience analysis model can distinguish the critical nodes in infrastructure systems and give the best and highly individualized approach to protect critical infrastructure systems. Overall, a new analysis model is developed for the protection of the critical infrastructure systems and therefore the targeted strategies and suggestions can be provided to protect the critical infrastructure systems.
    References | Related Articles | Metrics
    Application of Mobile App List in Evaluating Borrowers’ Credit Risk——An Empirical Analysis of an Online Lending Platform
    GUO Wei-dong, ZHOU Zach Zhi-zhong, QIAN Chun-tao
    2022, 30 (12):  96-107.  doi: 10.16381/j.cnki.issn1003-207x.2021.0359
    Abstract ( 331 )   PDF (1326KB) ( 424 )   Save
    With the development of the Internet and popularity of smart phones, data related to the use of mobile devices are used to study the default risk of borrowers, including communication records, short message receiving, mobile track and user behavior data.Data from a large online lending platform are adopted to study whether mobile Apps is related to the credit risk. Three types of Apps are analyzed, namely lifestyle Apps, financial Apps, and property Apps. Personal accounting Apps, takeout Apps and workout Apps are categorized as lifestyle Apps; fund Apps, stock Apps and future Apps are categorized as financial Apps; and car-buying Apps and house-buying Apps are categorized as property Apps. The empirical results show that the usage of these Apps is related to borrowers’ credit risk. Borrowers who install lifestyle Apps, financial Apps, and property Apps have significantly lower credit risk than those who do not install these Apps. In particular, accounting Apps, takeout Apps, stock Apps, and house-buying Apps are good indicators to identify borrowers with good credit.
    References | Related Articles | Metrics
    Research on Product Line Planning Considering Peer Influence in Social Network Context
    JIAO Yuan-yuan, YAN Xin, DU Jun, ROGER Jiao
    2022, 30 (12):  108-119.  doi: 10.16381/j.cnki.issn1003-207x.2021.1342
    Abstract ( 199 )   PDF (2175KB) ( 206 )   Save
    Under the personalized and diversified consumption trend, product line planning is an important way to meet consumer needs. However, in the traditional business environment, manufacturing companies lack appropriate “channels” to grasp the degree of consumer preference for various attributes of products, and they have a vague understanding of consumers’ product needs. With the widespread use of social networks, an opportunity is provided to solve the dilemma that consumers’ needs are difficult to identify. The reason is that it is possible to observe consumer demand for products by mining consumer browsing records, product reviews and other information in social networks. But it is a pity that although it has attracted the attention of the industry, the corresponding academic research is very scarce. In view of this, a master-slave joint optimization model (two-level programming model) covering the marketing and design levels is constructed by using Stackelberg game theory, prospect theory, and peer influence theory in the context of social networks. Among them, maximizing the number of consumers is regarded as the goal of marketing (master) by adjusting product attributes and attribute levels, and minimizing manufacturing costs is regarded as the goal of designing (slave) by adjusting product attributes and design parameter levels. Then, with the help of an example of smart phone product line planning, the nested gray wolf algorithm is used to numerically analyze the two-level planning model. And based on the results of numerical analysis, it is found that, first, the influence of peers in the context of social networks has positive significance on the number of consumers in the product line and its design and manufacturing costs. It not only contributes to the proliferation of products in social networks, but also contributes to the realization of product line design goals. Second, there is a positive correlation between the number of initial consumers and the number of consumers in the product line, and a negative correlation with manufacturing costs, and it has a “critical value” that can satisfy both the marketing level and the design level optimization goals. The value of this research lies in a joint optimization framework that couples social networks and product line planning is proposed. It not only helps to bridge the gap between engineering design and marketing, but also helps to solve the problem that consumer needs are difficult to identify in the traditional business environment. Of course, it provides a theoretical basis for product line planning for social networks.
    References | Related Articles | Metrics
    Epharmacy Demand Forecasting in the Presence of Promotional Activities
    LI Jian-bin, LEI Ming-hao, DAI Bin, CAI Xue-yuan
    2022, 30 (12):  120-130.  doi: 10.16381/j.cnki.issn1003-207x.2021.2661
    Abstract ( 375 )   PDF (1766KB) ( 680 )   Save
    E-pharmacy demand forecasting is highly affected by drug attributes and promotional activitiesproposed by the e-pharmacy platform. Atimeseries-machine learning hybrid model that integrates price discount and coupons is proposed to better analyze sales improvement brought by promotional activities, based on which more accurate forecasting results can be obtained. Traditional demand forecasting research decomposes demand under promotional activities into a linear combination of baseline sales and promotional lifting sales, while the drug’s treatment cycleis considered in this model, and SARIMA model is used to predict the baseline sales.Finally,predicted baseline sales data and promotional features are put into XGBoost model for integrated learning to further analyze the promotional effects. Sales data from a Chinese leading e-pharmacy is used to test the model’s effectiveness, results indicate that this proposed hybrid model performs better compared to the other three widely used forecasting models. At the same time, the hybrid model’s efficiency under different price discount, as well as promotional information and data pooling strategy is verified.Results show that the hybrid model performs better when price discount varies,promotional information can sufficiently reduce the forecasting error by at least 40% when is added into the proposed hybrid model, while data pooling strategy can help the hybrid model reduce forecasting error by around 10%. The proposed hybrid model is confirmed to be applicable and useful, which sheds light on e-pharmacy’s demand forecasting with promotional activities.
    References | Related Articles | Metrics
    Multi-objective Multi-learning Bacterial Foraging Optimization Algorithm for Mixed Data Clustering
    NIU Ben, , GUO Chen, TANG Heng
    2022, 30 (12):  131-140.  doi: 10.16381/j.cnki.issn1003-207x.2021.2694
    Abstract ( 234 )   PDF (2725KB) ( 338 )   Save
    With the easy generation and acquisition of data in medical, management, financial, and other fields, a large amount of data with mixed attributes is generated. How to mine valuable information from these kinds of data has attracted the attention of researchers. Clustering is one of the famous data mining methods, which can be employed to find information from the mixed attribute data sets. Various mixed-type data clustering methods have been designed, which can be divided into general clustering algorithms and evolutionary computation-based clustering algorithms. Among them, the evolutionary computation-based clustering algorithms mainly include single-objective or multi-objective optimization algorithms. These proposed algorithms show good performance under the specific context. However, when facing automatic clustering, high dimensional clustering, and multi-objective clustering problems, the algorithms in the first category cannot get satisfying clustering results; on the contrary, the algorithms in the second category show great potential. Therefore, the researchers have conducted in-depth research on the algorithms in the second category. When using the evolutionary computation-based clustering algorithms, two issues need to be taken into consideration further.On the one hand, these algorithms are proposed based on the K-prototype. It is well recognized that K-prototype employs the Hamming distance to compute the similarity of categorical attributes so that it cannot show the true relations between data samples. On the other hand, these algorithms mainly focus on the genetic algorithm, other evolutionary computation-based algorithms, such as bacterial foraging optimization algorithm, are worth studying in solving mixed-type data clustering problems.
    References | Related Articles | Metrics
    Environmental Regulation and Green Technological Innovation: Impact Mechanism Analysis and Spatial Spillover Effects
    OUYANG Xiao-ling, ZHANG Jun-hao, DU Gang
    2022, 30 (12):  141-151.  doi: 10.16381/j.cnki.issn1003-207x.2022.0642
    Abstract ( 387 )   PDF (1400KB) ( 548 )   Save
    How to achieve high-quality development with low carbon emission, one of the most important challenges for environmental regulation (ER) design, has attracted increasing attention from researchers and policymakers. On the basis of scientifically measuring the stringency of ER and the green technological innovation, the panel data of 274 cities at the prefecture level and above from 2005 to 2020 are used to examine the impact of ER on green technological progress and its mechanisms from the perspectives of heterogeneous effects and spatial spillovers.
    References | Related Articles | Metrics
    Manufacturer’s Information Acquisition Strategy under the Difference of Information Possession: Passive or Active
    TANG Min, ZHANG Zhao-qi, LI Zhi-guo
    2022, 30 (12):  152-161.  doi: 10.16381/j.cnki.issn1003-207x.2021.2686
    Abstract ( 273 )   PDF (1495KB) ( 229 )   Save
    With the arrival of the big data era and the intensification of global market uncertainty, more and more retailers use forecasting technology to forecast market demand. However, manufacturers who lack market information in the upper reaches of the supply chain often face asymmetric demand information relative to retailers, so they always need to weigh whether to implement the passive strategy or active strategy to obtain the demand information owned by downstream retailers. A supply chain consisting of a single manufacturer and a single retailer,the manufacturer pricing wholesale price and one-time transfer payment (such as brand license fee) is considered. The retailer sells the product to the consumer after wholesale from the manufacturer. The game model in the case of information symmetry, the information speculation, and the game model of information screening. The optimal information acquisition strategies of manufacturers and the effects of different strategies and retailers' information endowment advantages on the decision-making of manufacturers and suppliers are discussed through numerical simulation.
    References | Related Articles | Metrics
    The Forming Mechanism of Financial Asset Bubble Based on the Heterogeneous Price Beliefs
    HE Chao-lin, ZHANG Qi-xiang, CAO Wang-dong
    2022, 30 (12):  162-173.  doi: 10.16381/j.cnki.issn1003-207x.2021.0876
    Abstract ( 164 )   PDF (3047KB) ( 180 )   Save
    Financial asset bubble has been recurring phenomena in economic history, which has been observed in different time periods, in economies at different stages of development, and across a wide range of asset classes. It is also an anomaly of asset price, which strongly deviates from its fundamental value, has the devastating effects on financial market, results in the misallocation of resources, the impaired balance sheets, and etc., even the economic (financial) crisis. So, assuming that the future price changes of risky asset is formed by the extrapolation of its past price changes, a new asset price bubble model is proposed to study the forming mechanism of financial asset bubble, extract its intrinsic characteristics, and obtain relevant evidence for the stable development of financial market based on the cash-flow dividend shocks. Assuming that the investor of heterogeneous price beliefs has the preference of a constant absolute risk aversion utility, the optimal risky asset demand function of fundamental investor and extrapolative investor is obtained based on the model of expected utility; further, during the process of asset trading, it assumes that the extrapolative investor partly pays attention to the fundamental value of risky asset, modifies its optimal risky asset demand function, and obtains the asset price bubble model under the condition of market clearing; based on the setting of model parameter, it simulates the forming mechanism of financial asset bubble, and analyzes its inherent characteristics with the evidence from financial market; at last, based on the comparative setting of parameter value, it discusses the factors that affect the degree of strength and weakness of financial asset bubble from the perspective of investor’s heterogeneity and risk-free asset’s return. Results show, under the shocks of positive cash-flow dividend, the extrapolative trading behavior leads to the financial asset bubble, which has a lagging effect; financial asset bubble has the typical characteristic of three-stage, and the simultaneously rising of volume and price is a significant sign of the starting of financial asset bubble; the supply of risk-free asset is a suppressor of financial asset bubble; the difference of investor structure and their price beliefs is closely related to the degree of strength and weakness of financial asset bubble. The study not only provides an analytical framework for the forming mechanism of financial asset bubble, but also gives some relevant evidence for the stable development of financial market.
    References | Related Articles | Metrics
    Research on the Emission Reduction Effect of Green Finance from the Perspective of General Equilibrium Theory: Modeling and Empirical Test
    WEN Shu-yang, SHI Hao-ming, GUO Jian,
    2022, 30 (12):  173-184.  doi: 10.16381/j.cnki.issn1003-207x.2021.2630
    Abstract ( 313 )   PDF (2198KB) ( 561 )   Save
    Can green finance effectively promote carbon emission reduction? What is its internal mechanism? With the goal of carbon peaking and carbon neutrality, it is of great significance to answer this question scientifically and rigorously. The theoretical hypothesis is put forward that green finance can help carbon emission reduction by promoting the progress of green technology. This general equilibrium model with carbon emission constraints and endogenous emission reduction technology progress in the paper explains the internal mechanism of green finance to achieve carbon emission reduction by supporting technological progress, and provides empirical evidence based on China’s provincial panel data. The impact of green finance on carbon emission reduction from both theoretical and empirical dimensions is demonstrated, and it is pointed out that supporting technological innovation is an important mechanism for green finance to exert its emission reduction effect. Therefore, in the process of developing green finance, it is more effective to focus on the practical use of financial resources and effectively support the updating of emission reduction technologies than simply restricting the acquisition of financial resources in high-emission industries. At the same time, theoretical analysis shows that more green finance is not necessarily better, but there is an optimal scale. While actively promoting the development of green finance, it is still necessary to pay attention to overall planning to avoid unnecessary losses caused by financial “over-greening”.
    References | Related Articles | Metrics
    Patent Classification Based on multi-dimensional Feature and Graph Convolutional Networks
    WU Jie, GUI Liang, LIU Peng, SHENG Yong-xiang
    2022, 30 (12):  185-197.  doi: 10.16381/j.cnki.issn1003-207x.2021.2628
    Abstract ( 133 )   PDF (2264KB) ( 271 )   Save
    The shortening of patent examination time and the increase of patent number bring great challenges to patent classification, and using patent automatic classification technology to improve the efficiency of patent classification and shorten the time of patent examination has become an important research topic. An automatic patent classification framework is proposed based on multi-dimensional features and graph convolutional networks. The framework extracts the patent features from the dimensions of patent abstract, citation patent and patent inventor according to document metrology and graph representation learning theory. Secondly, the patent-core word network is constructed by using the dimensionality features of patent abstracts, and the dimensionality features of citation patents and patent inventors are embedded into the patent-core word network as patent number features. The semi-supervised learning of graph convolutional network is used to determine the classification labels of patent nodes in the patent-core word co-occurrence network and complete the task of patent automatic classification. In order to verify the effect of the method, the patent data from the Incopat global patent database are used for experiments. The experimental results show that the patent text information and the patent structured information as the patent features can improve the patent classification accuracy, and the introduction of backward citation patent information can improve the patent classification accuracy. At the same time, the framework proposed in this paper also provides a new solution to the problem of patent automatic classification, and provides support for the implementation of the policy of shortening patent examination time.
    References | Related Articles | Metrics
    Platform Supply Chain Risk Sharing Strategy under Procurement Empowerment
    XIAO Di, , ZHANG Xin-wei, PAN Ke-wen, , CONG Shuang
    2022, 30 (12):  198-210.  doi: 10.16381/j.cnki.issn1003-207x.2021.2624
    Abstract ( 225 )   PDF (2753KB) ( 284 )   Save
    In recent years, e-commerce platform companies have gradually become the leaders of the supply chain, and have empowered the upstream and downstream members of the platform supply chain from multiple dimensions such as procurement, brand, and data. Among them, procurement empowerment refers to the behavior that e-commerce platform companies integrate supply chain resources and various basic services with the help of digital intelligence technology to provide intelligent selection, centralized procurement and other procurement services for retailers, so as to reduce procurement costs, enhance procurement efficiency and improve the overall performance of supply chain. For example, Yunji selects a number of hot-selling products, integrates the orders of a large number of merchants, and uniformly purchases from high-quality brand suppliers. While ensuring product quality, it has resulted in a 30% reduction in total purchasing costs and a consumer repurchase rate of over 80%.
    References | Related Articles | Metrics
    Metacost Based Semi-supervised Heterogeneous Ensemble Model for Customer Credit Scoring
    YAN Lan, LI Si-han, XIAO Yi, KOU Yu-xuan, LIU Dun-hu, XIAO Jin
    2022, 30 (12):  211-221.  doi: 10.16381/j.cnki.issn1003-207x.2020.0933
    Abstract ( 168 )   PDF (2164KB) ( 293 )   Save
    With the popularization of the credit business, effective risk aversion is one of the main means to maintain stable profits in the financial industry, and credit risk is one of the most common and important risk types in the financial industry. Therefore, accurate credit scoring of customers is very important. However, the class distribution of customer data used for credit-scoring models is often highly imbalanced, which means that there are significantly more customers with good credit as compared to customers with bad credit, and only a few customers who have successfully obtained loans can be labeled according to their future behavior, many customers who have applied for loans but failed to obtain them cannot be labeled. These characteristics bring great challenges to the establishment of scientific and accurate customer credit-scoring models, and existing researches cannot solve the above problems well. To make up for the lack of existing researches, meta cost-sensitive learning, semi-supervised learning, and heterogeneous ensemble learning are combined, and a Metacost based semi-supervised heterogeneous ensemble model (Meta-Semi-HE) is proposed for customer credit scoring. This model includes the following three stages: 1) Metacost is used to modify the initial labeled training set to obtain Lm; 2) N heterogeneous classifiers hi(i=1,…, N) are trained on Lm by AdaBoost, concomitant ensemble Hi is used to selectively mark samples of unlabeled data set, and adds them into Lm, N heterogeneous classifiers are retrained with the new Lm. Repeat this step to improve the performance of the member classifiers until the termination condition is satisfied; 3) the final trained classifiers are used to classify samples of the test set. The empirical analysis is conducted in six customer credit-scoring datasets, and the results show that the Meta-Semi-HE has better customer credit-scoring performance than the other five models in the evaluation criteria of AUC, f, Type I accuracy, and Type II accuracy. A new way of thinking for banks’ customer credit-scoring modeling is provided, which helps banks to avoid risks more effectively and promotes the healthy and stable development of credit business in the financial industry.
    References | Related Articles | Metrics
    Macro Uncertainty and Crude Oil Price Risk Based on Nonparametric Multiple Expectile
    YAN Fu-lei, ZHANG Yu-tong, CUI Zhong-yue, ZHANG Gao-xun
    2022, 30 (12):  222-233.  doi: 10.16381/j.cnki.issn1003-207x.2021.2622
    Abstract ( 186 )   PDF (1863KB) ( 166 )   Save
    The crude oil price reflects the impact of macro uncertainty on economic activities and investors' expectations. Once the crude oil price is significantly impacted, the macro uncertainty will affect it in turn. Therefore, it is essential to construct a crude oil price risk prediction model that incorporates several macro variables simultaneously. The main models for measuring crude oil price risk include GARCH, APARCH, GARCH-MIDAS. However, the results may be biased due to the increasing number of parameters estimated by the above model after containing multiple variables.Thus, it aims to explore a risk prediction model that simultaneously contains several variables and has good prediction ability.The nonparametric multiple expectile is suitable for constructing the crude oil price risk measurement model embedded with multiple macro variables because it can estimate more than five explanatory variables.
    References | Related Articles | Metrics
    The Framework for Characteristic Factors of Poverty Statusby Using AI Algorithms:Related to the Path Choice of Rural Revitalization in China
    YUAN George Xianzhi, , , , , , ZHAO Min, LIU Hai-yang, ZHOU Yun-peng, YAN Cheng-xing, SHI Bao-feng, CHAI Na-na, LIN Jian-wu, HE Cheng-ying, MA Sheng, ZHANG Qian-you, DING Xiao-wei
    2022, 30 (12):  234-244.  doi: 10.16381/j.cnki.issn1003-207x.2021.2648
    Abstract ( 341 )   PDF (1862KB) ( 496 )   Save
    The goal of this paper is to establish a framework and associated analysis process for the extraction of related features to depict the poverty status of rural farmers. Based on the 18 types of data covered by a rural filing cardin China, combined with CART analysis and Gibbs sampling algorithm, 12 highly related characteristic factors are screened out to describe the poverty status of rural households.
    References | Related Articles | Metrics
    Prediction Method and Empirical Study of Precious Metal Futures Price
    CHEN Kai-jie, TANG Zhen-peng, WU Jun-chuang, ZHANG Ting-ting, DU Xiao-xu
    2022, 30 (12):  245-253.  doi: 10.16381/j.cnki.issn1003-207x.2020.0576
    Abstract ( 255 )   PDF (3187KB) ( 360 )   Save
    Accurate and reliable precious metal futures price forecasting is extremely crucial for investment decision-making and government gold reserves. In this paper, an adaptive model named VMD-Res.-EEMD-ELM is documented for predicting precious metal futures price, which combines the advantages of secondary decomposition and extreme learning machines. Variational modal decomposition (VMD) is selected as the main decomposition technique to generate a sequence of modal components (VMFi) and residual sequence (Res.). The ensemble empirical mode decomposition (EEMD) is used to perform secondary decomposition of the residual sequence. And then each component is put into the extreme learning machine (ELM) with good generalization ability to generate the outputs which will be superimposed to form the final prediction result. The proposed model not only makes full use of the advantages of the secondary decomposition technology, but also solves the problem that the traditional variational modal decomposition hybrid prediction model does not consider the influence of residuals. The new model is tested on the historical data of two daily return rate sequence - gold and silver futures price, which are collected from Choice financial terminal (the main financial data aggregator in China, an equivalent of Bloomberg). Empirical research exhibits that the hybrid model proposed in this paper can fully capture the characteristics of the daily return rate sequence of gold and silver futures price with the excellent performance which achieves directional accuracy(DSTAT) of 83.33% and 93.33%, and MAE of 0.15 and 0.11 respectively. Meanwhile, by comparison, the prediction accuracy of the proposed model is significantly higher than other existing models.
    References | Related Articles | Metrics
    Decisions of Hybrid ESupply Chain Considering Authentication Services and Consumer Preferences
    WAN Xiao-le, ZHOU Yuan, , QIN Man, LUO Jun-mei, ZHANG Kun-cheng,
    2022, 30 (12):  254-267.  doi: 10.16381/j.cnki.issn1003-207x.2021.0975
    Abstract ( 194 )   PDF (2994KB) ( 204 )   Save
    Authentication service platforms provide quality assurance for online commodity sales by offering authentication services, effectively improving consumer satisfaction and reducing the return rate of online products. In views of this, a hybrid E-supply chain system consisting of an authentication service E-supply chain and a traditional E-supply chain is constructed in this paper. Using Stackelberg game, the optimal pricing strategies and the corresponding profit under the centralized decision-making modeland the decentralized decision-making model were solved. The impact of authentication service efforts on the operation of each model is analyzed, and the optimal decisions of different models are compared. Finally, through numerical analysis, the model conclusions are verified. The research shows that: 1) Whether it is incentralized or decentralized decision-making model, the product sales prices in both channels increase with the growth of authentication se