Statistical Model Checking of the Keynes+Schumpeter Model: A Transient Sensitivity Analysis of a Macroeconomic ABM
This academic paper introduces a principled analysis framework for macroeconomic Agent-Based Models (ABMs) using Statistical Model Checking (SMC). Addressing the limitations of ad hoc Monte Carlo simulations, the authors demonstrate how SMC, implemented via the MultiVeStA tool, enables rigorous quantitative analysis without rewriting existing simulators. The study focuses on the heuristic-switching Keynes+Schumpeter (K+S) model, conducting a transient sensitivity analysis over a 600-step horizon. Key observables include unemployment, GDP growth, and market share. The methodology employs reusable temporal queries and confidence-based stopping rules to automatically determine necessary simulation effort for each parameter configuration. Results indicate that macro-financial and structural parameter sweeps generate significant transient effects, whereas heuristic-rule sweeps show weaker impacts under identical precision policies. This research highlights SMC's potential to enhance reproducibility in economic modeling by making uncertainty estimates and simulation costs explicit. The work bridges computer science techniques with economic theory, offering a robust method for analyzing complex, substantively rich economic systems.
Wire timeline
Statistical Model Checking of the Keynes+Schumpeter Model: A Transient Sensitivity Analysis of a Macroeconomic ABM
This academic paper introduces a principled analysis framework for macroeconomic Agent-Based Models (ABMs) using Statistical Model Checking (SMC). Addressing the limitations of ad hoc Monte Carlo simulations, the authors demonstrate how SMC, implemented via the MultiVeStA tool, enables rigorous quantitative analysis without rewriting existing simulators. The study focuses on the heuristic-switching Keynes+Schumpeter (K+S) model, conducting a transient sensitivity analysis over a 600-step horizon. Key observables include unemployment, GDP growth, and market share. The methodology employs reusable temporal queries and confidence-based stopping rules to automatically determine necessary simulation effort for each parameter configuration. Results indicate that macro-financial and structural parameter sweeps generate significant transient effects, whereas heuristic-rule sweeps show weaker impacts under identical precision policies. This research highlights SMC's potential to enhance reproducibility in economic modeling by making uncertainty estimates and simulation costs explicit. The work bridges computer science techniques with economic theory, offering a robust method for analyzing complex, substantively rich economic systems.
cs.AI updates on arXiv.org