Principal Model Analysis Based on Bagging PLS and PCA and Its Application in Financial Statement Fraud

Motivated by the Bagging Partial Least Squares(Bagging PLS)and Principal Component Analysis(PCA)algorithms,a novel approach known as Principal Model Analysis(PMA)method is introduced in this paper.In the proposed PMA algorithm,the PCA and the Bagging PLS are combined.In this method,multiple PLS models are trained on sub-training sets,derived from the training set using the random sampling with replacement approach.The regression coefficients of all the sub-PLS models are fused in a joint regression coefficient matrix.The final projection direction is then estimated by performing the PCA on the joint regression coefficient matrix.Subsequently,the proposed PMA method is compared with other traditional dimension reduction methods,such as PLS,Bagging PLS,Linear discriminant analysis(LDA)and PLS-LDA.Experimental results on six public datasets demonstrate that our proposed method consistently outperforms other approaches in terms of classification performance and exhibits greater stability.Additionally,it is employed in the application of financial statement fraud identification.PMA and other five algorithms are utilized to financial statement fraud which concerned by the academic community,and the results indicate that the classification of PMA surpassed that of the other methods.

A Superior Sales Mode for a Two-Echelon Supply Chain of Perishable Products

This paper considers sales mode selection issues for a two-echelon supply chain of perishable products involving a supplier and a retailer.Consumers’purchase desire is assumed to be positive correlation with the freshness of the product.First,the traditional sales mode is examined,in which the supplier and the retailer play a Stackelberg game.We propose a three-layer decision model for this situation,and obtain dynamic pricing strategies and selling cycle length.It is shown that the retailer has little motivation to order many perishable products so as to avoid a long selling cycle length.Second,the commission-charge mode is analyzed,in which the retailer declares its decision first.In this mode,we demonstrate that the perishable product will be on sale during the whole shelf life under a certain condition.The correlation between the sales price of each stage and the remaining shelf-life length is analyzed.Third,the superiority analysis for the two sales modes is conducted.We show the relation between the selling cycle lengths of the two modes.By our analysis,it is shown that both the supplier and the retailer gain more profits when the commission-charge mode is adopted and the commission rate locates in a certain open interval.Finally,a numerical illustration is presented to visualize the discussed models,and some supplements are made for the acquired conclusions by the illustration.

Particle Swarm Optimization Bat Algorithm Path Automatically Planning Research for Police Drones in Hilly Cities

Mountain cities are complex asymmetric dynamic network architectures,and the flight of UAVs in this environment is subject to various constraints,while efficiency is a crucial factor in the trajectory planning of police UAVs,which need to maintain high efficiency and safe flight paths between their starting and ending points,but the traditional trajectory planning method cannot meet the requirements of rapid maneuvering of police UAVs.To achieve this,a 3D terrain map is built,an objective function is established for the flight cost in the UAV trajectory planning process,and a planning algorithm called particle swarm optimization bat algorithm(PSOBA)is proposed.PSOBA combines the characteristics of the bat algorithm(BA)and the particle swarm optimization algorithm(PSO)to improve population diversity and resolve the delayed convergence issue in the last phases of BA.Simulation results show that PSOBA is more effective than BA,with a search time for the best solution that is approximately 20.43%shorter and a convergence value of the objective function that is approximately 38%smaller.PSOBA is also able to plan a quicker,shorter,and safer flight path compared to other trail planning algorithms that enhance the bat algorithm.These findings suggest that PSOBA is a powerful algorithm with potential application value in UAV trajectory planning control in the mobile intelligence era.Contribute to the service of public social security.

A Nonlinear Grey Bernoulli Model with Conformable Fractional-Order Accumulation and Its Application to the Gross Regional Product in the Cheng-Yu Area

This study considers a nonlinear grey Bernoulli forecasting model with conformable fractionalorder accumulation,abbreviated as CFNGBM(1,1,λ),to study the gross regional product in the ChengYu area.The new model contains three nonlinear parameters,the power exponentγ,the conformable fractional-orderαand the background valueλ,which increase the adjustability and flexibility of the CFNGBM(1,1,λ)model.Nonlinear parameters are determined by the moth flame optimization algorithm,which minimizes the mean absolute prediction percentage error.The CFNGBM(1,1,λ)model is applied to the gross regional product of 16 cities in the Cheng-Yu area,which are Chongqing,Chengdu,Mianyang,Leshan,Zigong,Deyang,Meishan,Luzhou,Suining,Neijiang,Nanchong,Guang’an,Yibin,Ya’an,Dazhou and Ziyang.With data from 2013 to 2021,several grey models are established and results show that the new model has higher accuracy in most cases.

A Method of Emergency Volunteer Team Internal Participation in Rescue Decision-Making Considering Psychological Behavior

A method for internal participation in rescue decision-making of emergency volunteer teams considering psychological behavior is proposed to address the time sequence of rescue tasks.Firstly,the problem of multi-tasking and multi-operation within the emergency volunteer team is described.Secondly,considering that task leaders are influenced by behavioral and psychological factors in the evaluation,the required time for the job is used as a reference point,and the expected time that volunteers can complete the job is used as an attribute value.The task leader's prospect satisfaction value for each volunteer is calculated based on prospect theory,and the perceived utility values of disappointment theory and regret theory are calculated to measure the task leader's satisfaction with each volunteer.Furthermore,a multilayer coded genetic algorithm is used to construct an optimization model for emergency volunteer decision-making with the objective of maximizing the satisfaction value.Finally,the feasibility and effectiveness of this method are illustrated by an example analysis.The result shows that the efficiency of rescue tasks can be improved through decision optimization within the volunteer team.

Should Low-Frequency-High-Consumption Enterprises Add Online-to-Offline Platforms?An Empirical Study Using the VAR Model

This study investigates the impact of online-to-offline(O2O)platforms,such as Ele.me and Meituan,on offline sales in low-frequency-high-consumption industries,specifically a mid-to-highend liquor distribution chain.Using data from 77 offline stores in Beijing collected during 2019-2022,the study employs a VAR model to analyze the relationship between offline sales and the use of O2O platforms.The results reveal a long-term equilibrium between the two,with most indicators showing a positive impact of O2O platforms on offline sales.The research provides valuable insights for lowfrequency-high-consumption enterprises in making multi-channel decisions and quantifies the impact of O2O platforms on offline sales.

What Drives Up Land Price in China?Evidence from Bidding Processes of Land Auctions in Beijing

The land price in big cities draws much attention and discussion for its skyrocketing appreciation.Most researches are from the macro perspective due to data restriction.This paper aims to investigate the critical factors in the price formation process of a land auction,using the listing auction micro bidding-level data in Beijing from 2013 to 2018.We construct a model for the relationship between quitting price and land,bidder's characteristics,housing market conditions and competitive intensity(including private and public signals),then we use OLS for identification.We find that competitive intensity increases the quitting price by causing competition and interaction between bidders.More importantly,we find evidence of cheating behavior in the land market.Results show that bidders have higher quitting prices when they are in a joint venture,and when a central SOE developer or a top 10 developer exist in the joint venture.We also find different behavior of developers in the short run and long run.Our research contributes to the literature of land auctions by analyzing the price formation process and developers'behavior.We also provide supporting evidence for the government to make adjustments of the auction system and identify the cheating developers.

Dynamic Risk Spillovers in Non-Ferrous Metal Future Markets During COVID-19:A Frequency Domain Analysis

During the COVID-19 pandemic,the international financial markets experienced severe turbulence.Under the background of“Made in China 2025”,substantial entity enterprises have a large demand for non-ferrous metals.With the enhancement of financial attributes of non-ferrous metals,it is vital to prevent financial systemic risk contagion in the non-ferrous metal markets.In this article,the ensemble empirical mode decomposition method is used to decompose the prices of eight important non-ferrous metals futures,and then the dynamic DY risk spillover index model is established from the perspectives of long-term and short-term.The risk spillover between non-ferrous metals during the COVID-19 is quantitatively analyzed from different frequency domains.The study finds that in the long run,the risk spillover relationship between non-ferrous metals remained basically stable,and the change of it after the epidemic is slight.In the short run,the risk spillover relationship has different degrees of structural changes after the outbreak of the COVID-19 pandemic.The ensemble empirical mode decomposition method can distinguish the risk spillovers in different cycles,and help to formulate policies for preventing systemic risks in the non-ferrous metal markets according to the different length of terms.

Improved Integrated Deep Model for Pap-Smear Cell Analysis

Cervical cancer is the fourth most common malignancy to strike a woman globally.If discovered early enough,it can be effectively treated.Although there is a chance of error owing to human error,the Pap smear is a good tool for first screening for cervical cancer.It also takes a lot of time and effort to complete.The aim of this study was to reduce the possibility of error by automating the process of classifying cervical cancer using Pap smear images.For the purpose of this study,dual convolution neural networks with LSTM were employed to classify images due to deep learning approaches inspire distinct features and powerful classifiers for many computer vision applications.The proposed deep learning model based on convolution neural networks(CNN)with the long short-term memory(LSTM)network is to learn features which give better recognition accuracy.The overall model is known as Smear-net.In which‘smear’indicates‘pap-smear cancer cells’and‘net’refers to neural network.The parameters such as,Accuracy,Precision,Recall,Accuracy,Sensitivity,and Specificity are used to validate the models.The proposed method provides the improved accuracy of 99.57 percentage for classification of the pap-smear cells.The proposed approaches demonstrate the effectiveness of our contributions by testing and comparing with the state-of-the-art techniques.

Zeroth-Order Methods for Online Distributed Optimization with Strongly Pseudoconvex Cost Functions

This paper studies an online distributed optimization problem over multi-agent systems.In this problem,the goal of agents is to cooperatively minimize the sum of locally dynamic cost functions.Different from most existing works on distributed optimization,here we consider the case where the cost function is strongly pseudoconvex and real gradients of objective functions are not available.To handle this problem,an online zeroth-order stochastic optimization algorithm involving the single-point gradient estimator is proposed.Under the algorithm,each agent only has access to the information associated with its own cost function and the estimate of the gradient,and exchange local state information with its immediate neighbors via a time-varying digraph.The performance of the algorithm is measured by the expectation of dynamic regret.Under mild assumptions on graphs,we prove that if the cumulative deviation of minimizer sequence grows within a certain rate,then the expectation of dynamic regret grows sublinearly.Finally,a simulation example is given to illustrate the validity of our results.

Market Rule or Equal Opportunity Rule:An Empirical Analysis Based on Acquisitions of Chinese Listed Companies

This paper examines the data of A-share listed companies in China from 2002 to 2017,drawing on the theory of equal opportunity and market rules in M&A transactions.This paper investigates the correlation between changes in tender offer policy and M&A tendencies and performance.The findings suggest that following the policy shift and the adoption of market rules,companies that secure an exemption from the mandatory tender offer obligation not only exhibit stronger M&A tendencies but also improved long-term M&A performance.This indicates that market rules are more suitable for China and contribute to enhancing the efficiency of the M&A market.The paper also presents evidence of a moderating effect,demonstrating that exemptions from the mandatory tender offer obligation positively influence the relationship between policy change and M&A performance.Lastly,this paper finds that state-owned and large-scale firms tend to exhibit a higher degree of M&A tendencies.

Analyst Coverage,Forecasting Bias,and Corporate Innovation:Evidence from China

Due to information asymmetry and strategic innovation,firms often encounter challenges related to insufficient driving forces and low-quality innovation outcomes.Analysts always act as information intermediaries who help foster the advancement of corporate innovation activities and the conversion of innovation output.This study examines the impact of analyst coverage and forecasting bias on corporate innovation,employing data from China A-shared listed firms spanning the period 2007 to 2019.We measure corporate innovation from two perspectives:Input and output.Specifically,we use the ratio of research and development(R&D)expenditure to sales as a proxy for the innovation input and the number of patent citations excluding self-citations to measure innovation output.We find that analyst coverage promotes corporate innovation,which is consistent with the“bright”side of analyst coverage.However,the positive effect of analyst coverage hinges on effectively transmitting and disclosing accurate information to investors in the capital market.Based on this,analysts'forecasting bias includes forecasting dispersion and optimism bias.We find evidence that an increase in analysts'forecast dispersion leads to a decrease in corporate innovation quality.Moreover,this paper presents a novel approach by employing the regression discontinuity method to examine the effect of analyst optimistic bias on firm innovation.The empirical findings reveal that overly optimistic forecasts by analysts exacerbate innovation quality.These analyses enrich the research on analyst coverage and corporate innovation,providing an empirical basis for improving the capital market with the help of analysts.

Research on Recommendation Algorithms Based on Cloud Models in Probabilistic Linguistic Environments

To solve the problem that the traditional cloud model can't directly process the textual review information in the recommendation algorithm,this paper combines the merits of the cloud model in transforming qualitative and quantitative knowledge with the multi-granularity advantages of probabilistic linguistic term sets in representing uncertain information,and proposes a recommendation algorithm based on cloud model in probabilistic language environment.Initially,this paper quantifies the attributes in the review text based on the probabilistic linguistic term set.Subsequently,the maximum deviation method is used to determine the weight of each attribute in the evaluation information of the product to be recommended,and the comprehensive evaluation number and attribute weight are converted into the digital characteristic value of the cloud model by using the backward cloud generator.Finally,the products are recommended and sorted based on the digital characteristic value of the cloud model.The algorithm is applied to the recommendation of 10 hotels,and the results show that the method is effective and practical,enriching the application of cloud models in the recommendation field.

How Does ESG Performance Impact Corporate Financing Costs?An Empirical Study in China

In the background of the green transformation of the economy and society,the ESG performance of enterprises has been paid more and more attention in the investment decision-making.However,previous studies have inadequately explored how the ESG performance affects corporate financing costs.Based on the information asymmetry theory,this paper analyzes the impact mechanism of ESG performance on corporate financing costs.Then,taking 1044 A-share listed companies in2016–2020 as a sample,through the sorting and analysis of ESG report disclosure and rating data,the company’s ESG performance indicators are obtained,and an empirical model is built to test the relationship between ESG performance and corporate financing costs.This paper constructs a panel regression model using ESG rating data and corporate financial data and finds that in the overall sample,the higher the ESG performance,the lower the equity financing cost;The higher the ESG performance,the lower the debt financing cost.In addition,it also discussed the moderating effect of enterprise scale and media attention on the impact of ESG performance on enterprise financing costs.The empirical results show that the influence of company size on ESG performance on financing costs has a moderating effect and a positive moderating effect.

A Novel Multi-Attribute Decision-Making Method Based on Probabilistic Hesitant Fuzzy Soft Set and Its Application

The probabilistic hesitant fuzzy multi-attribute group decision-making method introduces probability and hesitation into decision-making problems at the same time,which can improve the reliability and accuracy of decision-making results,and has become a research hotspots in recent years.However,there are still many problems,such as overly complex calculations and difficulty in obtaining probability data.Based on these,the paper proposes a multi-attribute group decision-making model based on probability hesitant fuzzy soft sets.Firstly,the definition of probabilistic hesitant fuzzy soft set is given.Then,based on soft set theory and probabilistic hesitant fuzzy set,the similarity measure of probabilistic hesitant fuzzy soft set is proposed,and the two measures are further combined.Finally,it is applied to the construction of multi-attribute group decision-making model,and the effectiveness and rationality of the model are verified by an example.The example shows that the new similarity calculation formula and algorithm model in this paper have higher accuracy,and the calculation process is more simple,it provides a feasible method for multi-attribute group decision making problems.

Research on Regional Differences and Influencing Factors of China's Carbon Emissions

This paper studies the regional differences,dynamic evolution and influencing factors of regional carbon emission intensity(CEI)in 262 cities and 5 regional urban agglomerations(UAs)in China.The Dagum Gini coefficient is used to analyze the intra-regional and inter-regional differences in carbon emissions,and the temporal evolution of the absolute differences of CEI among regions is analyzed by means of kernel density estimation(KDE).The paper provides an in-depth study on the spatial difference and temporal evolution of CEI in Chinese cities and major strategic regions.Through Moran index and LISA’s test,the spatial correlation of carbon emission in prefecture-level cities is tested,and its spatial agglomeration characteristics are described.It is found that China’s CEI is decreasing year by year,presenting a spatial pattern of“low in the south but high in the north”.Based on the calculation of carbon emission intensity at the urban level,this paper conducts LDMI factor decomposition research on carbon emission intensity at the national and key regions,and analyzes the impact of the impact factors on carbon emission intensity.The research results provide a path for China’s green development at the city level and urban agglomeration level,and a theoretical support for different regions and cities to introduce emission and carbon reduction policies.

A New Approach to University IT Project Portfolio Management Based on Multi-Criteria Methods and the COBIT 5 Governance Framework

Project portfolio management is a major challenge for some organizations.In most organizations,there are a large number of projects active at the same time,some not necessarily delivering value or not aligned with their strategic goals.Also universities face a lot of uncertainties when selecting and prioritizing the projects that make up their portfolio.In addition,the achievement of those who are aligned with the strategy of the university becomes a great challenge.So to ensure good project portfolio management,the implementation of selection and prioritization methods and processes becomes important.For the project portfolio management to be effective,it is necessary to establish a structured method adapted to the needs and strategy of the university.In this context,this paper proposes a method for selecting and prioritizing projects within the framework of the portfolio management dedicated to universities,which can promote harmony between the university’s strategy,the needs and the priority objectives for enable better decision-making.This method is based on the processes of the COBIT 5 good practice framework,and on the multi-criteria decision-making methods AHP,TOPSIS and the WSM technique,thus,it proposes seven project selection criteria based on the five axes IT governance strategies and two catalysts derived from COBIT 5 enablers.The evaluation and validation of this method was applied in the portfolio management of the Abdelmalek Essaadi Moroccan University(AUE).The result shows that this proposed method has made it possible to make a better selection and prioritization of the portfolio of projects of Abdelmleek Essaadi University having the most value.

Environmental Quality,Energy Consumption and Economic Inequality in China:Smooth Structural Shifts and Causal Linkages

The space-temporal evolution of economic inequality is examined with Markov chain test method,and the dynamic interrelationships among environmental quality,energy consumption,and economic inequality in China from the province-level are tested by focusing on accounting for structural shifts in causal linkages in this paper.We first employ the Toda-Yamamoto causality framework and then augment it with a Fourier approximation which captures structural changes as a smooth process.The empirical findings show that taking into account smooth structural shifts is important for the causal linkages between economic inequality and energy consumption,and also between environmental quality and energy consumption.The causality analysis with structural changes provides a causal linkage between economic inequality and energy consumption in 26 out 30 provinces and a causal linkage between environmental quality and energy consumption in 7 out 30 provinces,while the quantities are 22 out 30 and 5 out 30 respectively when not accounting for structural shifts.These findings are consistent with the fact that provincial economics in China have experienced structural changes in economy-environment-energy sectors.We also conduct additional analyses which point out that regional and cyclical dependency matter for the causal relationships,and the method of HP filtering can not effectively solve the problem of smooth shifts in economy-environment-energy causality.

Predicting Critically Ill Patients Short-Term Mortality Risk Using Routinely Collected Data:Deep Learning Model Development,Validation,and Explanation

This paper aims to develop and validate a deep learning-based short-term mortality risk prediction model for critically ill patients by using routinely collected data in a large Chinese cohort and explore the explainability of the model decision.A total of 10925 critically ill patients between January 2014 and June 2020 are included in this study.Data routinely collected in the electronic health records(EHRs)system are extracted and used to develop a short-term mortality risk prediction model based on a deep artificial neural network(ANN).The features include demographic characteristics,vital signs,laboratory tests,and the daily dose of intravenous medications.The developed deep learning model(AUROC:0.88,AUPRC:0.63,Brier score:0.108)is superior to the model based on APACHEⅡscores(AUROC:0.78,AURPC:0.52,Brier score:0.124)in the prediction of hospital mortality for critically ill patients.Further attribution analysis based on the integrated gradients method shows that measurements observed at a later time seem to have a more significant influence on mortality,while earlier usage of amiodarone or dexmedetomidine contributed to lower mortality.This well-performing and interpretable model may have practical implications for improving the quality of care for critically ill patients.

Parallelization and Acceleration of Dynamic Option Pricing Models on GPU-CPU Heterogeneous Systems

In this paper,stochastic global optimization algorithms,specifically,genetic algorithm and simulated annealing are used for the problem of calibrating the dynamic option pricing model under stochastic volatility to market prices by adopting a hybrid programming approach.The performance of this dynamic option pricing model under the obtained optimal parameters is also discussed.To enhance the model throughput and reduce latency,a heterogeneous hybrid programming approach on GPU was adopted which emphasized a data-parallel implementation of the dynamic option pricing model on a GPU-based system.Kernel offloading to the GPU of the compute-intensive segments of the pricing algorithms was done in OpenCL.The GPU approach was found to significantly reduce latency by an optimum of 541 times faster than a parallel implementation approach on the CPU,reducing the computation time from 46.24 minutes to 5.12 seconds.