economic impact modeling

Conventional optimization practices frequently overlook externalities and their feedback loops within socio-economic systems, leading to decisions characterized by ignorance, misjudgment, and short-termism. Method: This paper introduces an integrative framework that unifies systems thinking with externality economics—achieving, for the first time, deep coupling of normativity (value trade-offs and responsibility calibration), dynamics (feedback-loop modeling), and quantifiability (shadow pricing and social cost accounting). It employs system dynamics modeling, stakeholder mapping, and structured externality assessment to systematically identify affected parties, clarify externality transmission mechanisms, and specify *when* and *how* externalities should be incorporated into optimization processes. Contribution/Results: The framework delivers actionable pathways for embedding optimization in algorithmic governance, public policy, and AI ethics, alongside methods for responsibility calibration. It overcomes key limitations of traditional optimization—its neglect of interconnectivity, dynamic feedback, and pluralistic value structures.

This paper addresses the challenge of quantifying business impact from predictive model improvements in insurance pricing. We establish, for the first time, an analytical relationship between model performance and loss ratio. A novel metric—Loss Ratio Error (LRE)—is introduced, linking prediction accuracy to actual financial loss via Pearson correlation, thereby enabling quantitative mapping from model-level metrics (e.g., RMSE) to business-level KPIs. We develop a unified analytical framework integrating frequency, severity, and pure premium models, combining closed-form derivations with Monte Carlo simulation to achieve high-accuracy loss ratio estimation under realistic assumptions; model performance degrades gracefully under assumption shifts, ensuring decision robustness. Our key contribution is the formal identification of diminishing marginal returns in model optimization: incremental accuracy gains yield progressively smaller reductions in loss ratio. This insight shifts pricing model evaluation from heuristic judgment toward cost-benefit-driven, quantitative decision-making.

This paper addresses the challenge of jointly estimating multiple causal effects—such as multidimensional price response curves—from observational data alone, under incentive strategy optimization, to enhance modeling robustness and decision effectiveness in distributionally anomalous settings. Method: We propose the first deep monotonic neural network framework capable of simultaneously modeling multiple causal effects, integrating monotonicity constraints, multi-task learning, and counterfactual estimation. We further introduce three empirically verifiable prior conditions that theoretically characterize the applicability boundary of observational data for causal evaluation. Contribution/Results: Evaluated via offline benchmarking and online A/B testing, our model achieves significant improvements in prediction accuracy and strategy ROI. The results empirically validate both the feasibility and necessity of joint multi-causal-effect modeling without randomized controlled trials (RCTs), advancing causal inference for real-world incentive design.

This paper addresses the challenge of accurately identifying demand under substantial temporal fluctuations and absent cost-variation information, focusing on the French railway industry. We systematically evaluate the economic performance of revenue management (RM) strategies using a novel identification framework that integrates time-series relative price changes, consumer rational expectations, and firms’ weak optimality conditions in pricing. Our methodology combines structural econometric modeling, counterfactual demand estimation, endogenous price treatment, and censoring-handling techniques to overcome identification issues arising from sales cutoffs and the lack of exogenous price variation. Results show that current RM practices significantly outperform uniform pricing but still incur a 16.7% revenue loss relative to theoretically optimal dynamic pricing. This study provides the first empirical quantification of RM’s net economic value in a real-world industrial setting and reveals its critical role in aggregating and processing information under demand uncertainty.

This paper identifies a fundamental bias in estimating average treatment effects (ATE) in continuous-parameter A/B tests—particularly price experiments—arising from interference among market participants. In pricing contexts, conventional estimators of profit change expectations can exhibit sign reversal, leading firms to adopt profit-damaging pricing policies. To address this, we propose a lightweight debiasing method requiring only equal partitioning of experimental units. We are the first to systematically characterize the “sign reversal” phenomenon and prove its ubiquity in two-sided markets and multi-category commission pricing. Through structural modeling and differential analysis, we derive an explicit closed-form expression for the bias and theoretically demonstrate that the classical estimator can indeed flip sign. Empirical evaluations across diverse market settings confirm that our method consistently restores correct decision directionality, thereby ensuring reliable causal inference.

This study investigates how policy interventions can balance consumer surplus and firm profits in the AI supply chain to prevent excessive extraction of consumer welfare by upstream foundation model providers and downstream application firms. The authors develop a game-theoretic model featuring an upstream provider and two competing downstream firms, incorporating compute costs and data preprocessing costs to systematically evaluate the economic effects of price competition, quality competition, and compute subsidies. The analysis delineates the effectiveness boundaries of various regulatory policies under different cost structures, proposes a complementary policy mechanism, and identifies conditions for achieving a tripartite win-win among consumers, upstream providers, and downstream firms: quality competition consistently enhances consumer surplus, while price competition and compute subsidies exhibit complementary efficacy across high- and low-cost scenarios.

Traditional tourism demand forecasting often treats accommodation supply as exogenous and fixed, overlooking the elasticity, endogeneity, and policy responsiveness of supply in platform-based short-term rental markets, which leads to model failure under supply fluctuations. This study proposes a supply-demand coupled forecasting framework that explicitly models the dynamic interaction between supply and demand through three dimensions: agent behavior, information dynamics, and policy interventions, thereby addressing the observational truncation caused by inventory sell-outs. For the first time, it systematically incorporates supply endogeneity into tourism demand prediction by integrating revenue management principles, two-sided market theory, and Bayesian time series methods. Through simulation, the work reveals the identification bias inherent in conventional models and establishes a new paradigm for jointly forecasting supply and demand.

This study addresses the unresolved trade-off between investing in predictive capabilities and alternative policy instruments—such as capacity expansion or service quality improvements—in resource-scarce allocation settings. The authors propose an empirical framework integrating causal inference, counterfactual simulation, and welfare economics, and introduce rvp, the first operational open-source toolkit for quantifying the marginal welfare effects of prediction in resource allocation and enabling cross-context policy comparisons. The framework’s validity is demonstrated through two empirical applications: job placement services in Germany and poverty targeting in Ethiopia. Results reveal that the welfare value of prediction is highly context-dependent, offering policymakers a scalable benchmark for evaluating and prioritizing interventions under constrained resources.

Estimating market impact requires reconstructing counterfactual price paths for unexecuted trades under shared sources of randomness, yet these paths are inherently unobservable. This work proposes the first exact conditional simulation method for history-dependent marked point processes, leveraging a sparse representation of Poisson random measures to enable event-driven reconstruction of counterfactual trajectories under perturbed intensities. The approach yields unbiased and consistent path-level estimates of market impact for aggressive, passive, and hybrid trading strategies alike, establishing a rigorous theoretical foundation and an efficient computational framework for evaluating trading impact in high-frequency settings.

This study investigates the spillover effects of corporate toxic emissions and their network transmission mechanisms. Leveraging a panel dataset of U.S. industrial facilities from 2000 to 2023, the paper eschews pre-specified network structures and instead endogenously identifies emission impact networks through high-dimensional data, constructing a data-driven spatial panel model to flexibly estimate both direct and indirect facility-level effects. The findings reveal that approximately 28% of the total emission effect stems from indirect spillovers—substantially higher than estimates derived from conventional network specifications based on geographic proximity or industry classification—thereby exposing systematic biases in those traditional approaches. This methodology offers a more reliable network foundation for environmental risk assessment and targeted regulatory interventions.