Portail HAL Rennes SB
Not a member yet
692 research outputs found
Sort by
The role of promotion versus prevention-orientation to predict individual cryptocurrency participation
No full textInternational audienc
Optimal unemployment insurance in a THANK model
No full textInternational audienceWe characterize optimal unemployment insurance (UI) in a heterogeneous-agent model with unemployment risk and sticky prices. In the long run, the optimal reform calls for a lower replacement rate that raises vacancies and lowers unemployment. In the short run, the optimal reform raises the replacement rate initially to smooth real wage adjustments along the transition and attenuate short-run welfare losses. Once at its optimal level, the replacement rate should vary counter-cyclically in response to demand shocks. Productivity shocks generate quasi-efficient fluctuations and call for a quasi-constant replacement rate. The aggregate welfare gains from an optimal reform are large, around 1% of equivalent consumption. The aggregate welfare gains from an optimal UI policy over the business cycle are smaller, around 0.2%, and essentially vanish with flexible prices because the aggregate demand stabilization motive is muted
How Do Coalitions Break Down? An Alternative View
No full textInternational audienceWe propose an alternative dynamic theory of coalition breakdown. Motivated by recent coalition-splitting events of unilateral country withdrawals, we assume that: (i) the payoff-sharing rule within coalitions is not necessarily set according to any optimality and/or stability criterion and (ii) players initially behave as if the coalition will last forever. If the sharing rule is non-negotiable or if renegotiation is very costly, compliance with these rules may become unbearable for a given member because the rule, being too rigid, would make exit preferable as time passes. We examine this endogenous exit problem in the case of time-invariant sharing rules. Assuming a Nash non-cooperative game after a (potential) split where players play Markovian strategies, we characterize the solutions of the endogenous exit problem in a linear-quadratic frame with endogenous splitting time. We find that splitting countries are precisely those that used to benefit the most from the coalition. Suitable sharing rules should be used to prevent coalition splitting. When initial pollution is high, all shares should be low enough and none of the players should receive a payoff share larger than 1/2. If initial pollution is low, we provide an explicit interval for sharing-rule values to prevent the collapse of the coalition. Finally, we demonstrate that the latter properties are qualitatively consistent with the optimal behaviors and equilibrium outcomes resulting from players anticipating the end of the coalition and acting accordingly
Integration and substitution in hybrid manufacturing and refurbishing systems
No full textInternational audienc
Equilibrium anti-counterfeiting strategies with deceptive counterfeits: Proactive, reactive, or instantaneous?
No full textInternational audienceThis papers studies anti-counterfeiting strategies with deceptive counterfeits. We develop a model to study the interactive anti-counterfeiting and counterfeit hiding decisions between the genuine brand company and a deceptive counterfeiter. Facing deceptive counterfeits, the genuine brand company can choose to adopt a proactive, instantaneous, or reactive anti-counterfeit strategy. We first examine these strategies and then characterize the equilibrium outcomes. Surprisingly, we reveal that the genuine brand company does not necessarily benefit from heavier penalty and the counterfeiter is not necessarily worse off with more costly counterfeit hiding efforts. Interestingly, the counterfeiter’s hiding effort decreases with heavier penalty when the genuine brand company’s anti-counterfeit and the counterfeiter’s hiding decisions are sufficiently efficient. Besides, higher levels of counterfeit imitation or penetration can either hurt or benefit the genuine brand company. Whether the counterfeiter is better off with deeper counterfeit penetration depends on the status quo. Furthermore, the counterfeiter exerts fewer counterfeit hiding efforts when the genuine brand company exerts more anti-counterfeit efforts, while the anti-counterfeit effort increases with the counterfeit hiding effort, regardless of the anti-counterfeiting strategies. Finally, for decision makers and policy makers, we discuss the implications for anti-counterfeiting strategies in practice
A risk-averse distributionally robust project scheduling model to address payment delays
No full textInternational audienceDelays in payments have become a common risk factor for industrial projects, especially in recent years, since the financial position of firms has been threatened by pandemics, wars, inflation, and major supply chain disruptions. These delays create a time lag between expenses and payments, potentially leading to cash shortages that can have significant negative effects on the project success. To address cash shortage issues, project contractors often explore alternative financing options. The amount of money the contractor needs to borrow and when the loan is taken out considerably affects the overall project cost. In this paper, we present a distributionally robust model for effective cash flow management that minimizes the financing cost by accurately estimating the amount and timing of the expenses and revenues throughout the project life cycle. For the proposed model, we develop a heuristic algorithm that solves the problem efficiently. The performance of the heuristic is compared to the best-known solutions generated within a time limit by an off-the-shelf exact solver. Our results show that our algorithm is very competitive and can generate better solutions in substantially less time
Economic epidemiological modelling: A progress report
No full textInternational audiencePrior to the Covid-19 crisis, the integration of epidemiology and economics that is, economic epidemiology modelling (epi-econ), was relatively limited. The emergence of the Covid-19 crisis has prompted an unprecedented surge in this literature. This paper identifies and develops the main conceptual and modelling challenges involved in the expanding epi-econ stream, with a particular attention to the mathematical issues due, in particular, to the non-convex nature of epi-econ models. Recent extensions are also examined and a few future areas of research highlighted
How do organizations leverage social media to enhance marketing performance? Unveiling the power of social CRM capability and guanxi
No full textInternational audienceDrawing on information technology-enabled organizational capabilities theory, we argue that how and when social media use affects marketing performance depending on the mediating role of social customer relationship management (CRM) capability and the moderating role of guanxi. Using a sample of 194 Chinese agricultural firms, we employed a combination of complementary techniques: advanced analysis for composites–structural equation modeling (ADANCO-SEM) and fuzzy-set qualitative comparative analysis (fsQCA). In ADANCO-SEM analysis, we empirically verify the full mediating role of social CRM capability in the association between social media use and marketing performance and identify the condition for this indirect effect (i.e., when the guanxi levels are high). Moreover, guanxi plays a dual moderating role—it exerts a substitution effect when social media is used to build social CRM capability, and it exerts a complementary effect when social CRM capability is leveraged to improve marketing performance. FsQCA analysis further reveals the different configurations of social media use, guanxi, and social CRM capability that can lead to high marketing performance in different firm sizes. The findings complement those from ADANCO-SEM analysis vis-à-vis the conditions for the formation of marketing performance. This study contributes to the emerging literature on the business value of social media by theoretically exploring and empirically validating the mechanisms and boundary conditions by which social media affects marketing performance
Convergence or Divergence: A Computational Text Analysis of Stakeholder Concerns on Manufacturing Upgrading in China
No full textInternational audienceStakeholder participation is essential to the reasonable design and smooth implementation of industrial policies. Discourse analysis can be employed as a valuable methodology to decode stakeholder (dis)agreements. Following stakeholder theory, we examine four stakeholder groups, namely the government, enterprises, media, and academia, to analyze the public focus of industrial upgrading in China. We adopt a computational text analysis approach (keyword frequency calculation, word collocation analysis, and theme identification) to understand the divergent and convergent stakeholders’ concerns toward manufacturing upgrading in China from 2015 to 2020, which is widely considered as a manufacturing upgrading policy formulation stage. Our results show that stakeholders mainly focus on innovation capability and digital transformation for China's manufacturing upgrading. There, industrial planning is the major issue for government and academia. Contradistinctively, enterprises’ main concern is servitization. Enterprise internationalization is more frequently mentioned than manufacturing upgrading in the media industry. Policymakers should engage in various tactics to make the policies endorsed by other stakeholders. Most importantly, the government should fully integrate the views of entrepreneurs in the policy initiation stage. We provide practical implications for the Chinese government to implement better the Made in China 2025 plan
Explainable artificial intelligence in transport Logistics: Risk analysis for road accidents
No full textInternational audienceAutomobile traffic accidents represent a significant threat to global public safety, resulting in numerous injuries and fatalities annually. This paper introduces a comprehensive, explainable artificial intelligence (XAI) artifact design, integrating accident data for utilization by diverse stakeholders and decision-makers. It proposes responsible, explanatory, and interpretable models with a systems-level taxonomy categorizing aspects of driver-related behaviors associated with varying injury severity levels, thereby contributing theoretically to explainable analytics. In the initial phase, we employed various advanced techniques such as data missing at random (MAR) with Bayesian dynamic conditional imputation for addressing missing records, synthetic minority oversampling technique for data imbalance issues, and categorical boosting (CatBoost) combined with SHapley Additive exPlanations (SHAP) for determining and analyzing the importance and dependence of risk factors on injury severity. Additionally, exploratory feature analysis was conducted to uncover hidden spatiotemporal elements influencing traffic accidents and injury severity levels. We developed several predictive models in the second phase, including eXtreme Gradient Boosting (XGBoost), random forest (RF), deep neural networks (DNN), and fine-tuned parameters. Using the SHAP approach, we employed model-agnostic interpretation techniques to separate explanations from models. In the final phase, we provided an analysis and summary of the system-level taxonomy across feature categories. This involved classifying crash data into high-level causal factors using aggregate SHAP scores, illustrating how each risk factor contributes to different injury severity levels