Government intervention model based on behavioral heterogeneity for China’s stock market

Active government intervention is a striking characteristic of the Chinese stock market.This study develops a behavioral heterogeneous agent model(HAM)comprising fundamentalists,chartists,and stabilizers to investigate investors’dynamic switching mechanisms under government intervention.The model introduces a new player,the stabilizer,into the HAM as a proxy for the government.We use the model to examine government programs during the 2015 China stock market crash and find that it can replicate the dynamics of investor sentiment and asset prices.In addition,our analysis of two simulations,specifically the data-generating processes and shock response analysis,further corroborates the key conclusion that our intervention model not only maintains market stability but also promotes the return of risk asset prices to their fun-damental values.The study concludes that government interventions guided by the new HAM can alleviate the dilemma between reducing price volatility and improving price efficiency in future intervention programs.

Modeling heterogeneous behaviors with different strategies in a terrorist attack

Existing simulations of terrorist attacks do not consider individual variations.To overcome this lim-itation,we propose a framework to model heterogeneous behavior of individuals during terrorist attacks.We constructed an emotional model that integrated personality and visual perception for pedestrians.The emotional model was then integrated with pedestrian relationship networks to establish a decision-making model that sup-ported pedestrians’altruistic behaviors.A mapping model has been developed to correlate antisocial personality traits with attack strategies employed by terrorists.Experiments demonstrate that the proposed algorithm can generate practical heterogeneous behaviors that align with existing psychological research findings.

Capturing driving behavior Heterogeneity based on trajectory data

Driving behavior is heterogeneous for various drivers due to the different influencing factors as reaction time,gender,driving years and so on.Some existing works tried to reproduce some of the complex characteristics of real traffic flow by taking into account the heterogeneous driving behavior,and the drivers are generally divided into two classes(including aggressive drivers and careful drivers)or three classes(including aggressive drivers,normal drivers and careful drivers).Nevertheless,the classification approaches have not been verified,and the rationality of the classifications has not been confirmed as well.In this study,the trajectory data of drivers is extracted from the NGSIM datasets.By combining the K-Means method and Silhouette measure index,the drivers are classified into four clusters(named as clusters A,B,C and D,respectively)in accordance with the acceleration and time headway.The two-dimensional approach is applied to analyze the characteristics of different clusters.Here,one dimension consists of“Cautious”and“Aggressive”behaviors in terms of velocity and acceleration,and the other dimension consists of“Sensitive”and“Insensitive”behaviors in terms of reaction time.Finally,the fuel consumption and emissions for different clusters are calculated by using the VT-Micro model.A surprising result indicates that overly“cautious”and“sensitive”behaviors may result in more fuel consumption and emissions.Therefore,it is necessary to find the balance between the driving characteristics.

Heterogeneous influence of individuals’ behavior on mask efficacy in gathering environments

Wearing masks is an easy way to operate and popular measure for preventing epidemics.Although masks can slow down the spread of viruses,their efficacy in gathering environments involving heterogeneous person-to-person contacts remains unknown.Therefore,we aim to investigate the epidemic prevention effect of masks in different real-life gathering environments.This study uses four real interpersonal contact datasets to construct four empirical networks to represent four gathering environments.The transmission of COVID-19 is simulated using the Monte Carlo simulation method.The heterogeneity of individuals can cause mask efficacy in a specific gathering environment to be different from the baseline efficacy in general society.Furthermore,the heterogeneity of gathering environments causes the epidemic prevention effect of masks to differ.Wearing masks can greatly reduce the probability of clustered epidemics and the infection scale in primary schools,high schools,and hospitals.However,the use of masks alone in primary schools and hospitals cannot control outbreaks.In high schools with social distancing between classes and in workplaces where the interpersonal contact is relatively sparse,masks can meet the need for prevention.Given the heterogeneity of individual behavior,if individuals who are more active in terms of interpersonal contact are prioritized for mask-wearing,the epidemic prevention effect of masks can be improved.Finally,asymptomatic infection has varying effects on the prevention effect of masks in different environments.The effect can be weakened or eliminated by increasing the usage rate of masks in high schools and workplaces.However,the effect on primary schools and hospitals cannot be weakened.This study contributes to the accurate evaluation of mask efficacy in various gathering environments to provide scientific guidance for epidemic prevention.

Multi-scale study on the heterogeneous deformation behavior in duplex stainless steel

The heterogeneous deformation behavior of austenite and ferrite in the 2205 duplex stainless steel was subjected to multiscale analysis based on the in situ synchrotron-based high energy X-ray diffraction,microscopic digital image correlation,electron backscatter diffraction,and transmission electron microscopy.It is found that the heterogeneous deformation triggers from the yielding of austenite.During this deformation stage,austenite experiences greater strain in the area near the phase boundaries because of the impeded function of the phase boundaries to dislocations.Owing to the relatively small difference in hardness between the constituent phases,the strain in austenite grains extends into the adjacent ferrite grains when entering into the ferrite yielding stage.In addition,the strain distribution of the austenite grains is more homogeneous than that of the ferrite grains because of the lower stacking fault energy of austenite,which results in a planar slip,and higher stacking fault energy in case of ferrite,causing cross slip.The interaction between austenite and ferrite becomes considerably obvious when the strain further increases after both constituent phases yielding because of the back stress and forward stress in austenite and ferrite,respectively,which are generated by the pile-up of the geometrically necessary dislocations.

BGNN: Behavior-aware graph neural network for heterogeneous session-based recommendation

Session-based recommendation(SBR)and multibehavior recommendation(MBR)are both important problems and have attracted the attention of many researchers and practitioners.Different from SBR that solely uses one single type of behavior sequences and MBR that neglects sequential dynamics,heterogeneous SBR(HSBR)that exploits different types of behavioral information(e.g.,examinations like clicks or browses,purchases,adds-to-carts and adds-to-favorites)in sequences is more consistent with real-world recommendation scenarios,but it is rarely studied.Early efforts towards HSBR focus on distinguishing different types of behaviors or exploiting homogeneous behavior transitions in a sequence with the same type of behaviors.However,all the existing solutions for HSBR do not exploit the rich heterogeneous behavior transitions in an explicit way and thus may fail to capture the semantic relations between different types of behaviors.However,all the existing solutions for HSBR do not model the rich heterogeneous behavior transitions in the form of graphs and thus may fail to capture the semantic relations between different types of behaviors.The limitation hinders the development of HSBR and results in unsatisfactory performance.As a response,we propose a novel behavior-aware graph neural network(BGNN)for HSBR.Our BGNN adopts a dual-channel learning strategy for differentiated modeling of two different types of behavior sequences in a session.Moreover,our BGNN integrates the information of both homogeneous behavior transitions and heterogeneous behavior transitions in a unified way.We then conduct extensive empirical studies on three real-world datasets,and find that our BGNN outperforms the best baseline by 21.87%,18.49%,and 37.16%on average correspondingly.A series of further experiments and visualization studies demonstrate the rationality and effectiveness of our BGNN.An exploratory study on extending our BGNN to handle more than two types of behaviors show that our BGNN can easily and effectively be extended to multibehavior scenarios.

Exploring financially constrained small- and medium-sized enterprises based on a multi-relation translational graph attention network

Financing needs exploration(FNE),which exploresfinancially constrained small-and medium-sized enterprises(SMEs),has become increasingly important in industry forfinancial institutions to facilitate SMEs’development.In this paper,wefirst perform an insightful exploratory analysis to exploit the transfer phenomenon offinancing needs among SMEs,which motivates us to fully exploit the multi-relation enterprise social network for boosting the effectiveness of FNE.The main challenge lies in modeling two kinds of heterogeneity,i.e.,transfer heterogeneity and SMEs’behavior heterogeneity,under different relation types simultaneously.To address these challenges,we propose a graph neural network named Multi-relation tRanslatIonal GrapH a Ttention network(M-RIGHT),which not only models the transfer heterogeneity offinancing needs along different relation types based on a novel entity–relation composition operator but also enables heterogeneous SMEs’representations based on a translation mechanism on relational hyperplanes to distinguish SMEs’heterogeneous behaviors under different relation types.Extensive experiments on two large-scale real-world datasets demonstrate M-RIGHT’s superiority over the state-of-the-art methods in the FNE task.