general equilibrium modeling

This study addresses the ambiguity in policy analysis arising from the lack of a unified equilibrium selection mechanism in dynamic stochastic general equilibrium (DSGE) models under multiple equilibria. Viewing DSGE models as fixed-point selection devices within self-referential economies, the paper proposes a unified framework comprising model specification, a self-reference operator, and a programmable equilibrium selector. It formalizes equilibrium selection as a computable operation for the first time, demonstrating that the Blanchard–Kahn conditions correspond to a specific selector and introducing alternative rules—such as minimum variance and fiscal anchoring—to better reflect policy intent. Leveraging linear rational expectations systems and standard solution techniques like QZ decomposition and OccBin, the framework enables efficient computation and validation of selectors. This approach reinterprets mainstream DSGE solution methods, facilitates systematic comparison of selection rules, and significantly enhances model transparency and policy relevance.

This paper addresses estimation and inference in macroeconomic models with multiple behavioral equilibria when agents form expectations via constant-gain learning rules. Facing challenges—including mixed convergence rates of equilibrium paths and nonstandard limiting distributions—we propose a structural parameter estimator based on nonlinear least squares and rigorously establish its strong consistency and asymptotic normality. We further construct uniform confidence bands for equilibrium paths and develop a novel statistical inference framework tailored to recurrent solution scenarios. Our methodology integrates geometric ergodicity analysis, analytical derivation, and Monte Carlo simulation. Empirical and simulation results demonstrate excellent finite-sample statistical properties and robustness. The proposed approach provides a practical, theoretically grounded tool for empirical identification of learning-based macroeconomic models.

Rational expectations in heterogeneous-agent macroeconomic models induce a “curse of dimensionality”: equilibrium price predictions require the entire cross-sectional distribution as a state variable, rendering the Bellman equation infinite-dimensional and computationally and cognitively intractable—severely limiting applicability to aggregate risk and nonlinear crisis analysis. Method: We systematically critique the theoretical and empirical deficiencies of rational expectations and propose an alternative expectation paradigm grounded in three criteria: computational feasibility, empirical consistency, and partial immunity to the Lucas critique. This framework integrates temporary equilibrium, survey-based expectations, least-squares learning, and reinforcement learning. We employ dynamic programming, master equation modeling, and behavioral–machine learning hybrid methods. Contribution/Results: We formally establish the fundamental infeasibility of rational expectations in high-dimensional heterogeneous environments and deliver the first analytically rigorous yet numerically tractable framework for heterogeneous-agent macroeconomics—opening a viable path for modeling systemic risk and crises.

This paper addresses the challenge of building a verifiable, heterogeneous-agent macroeconomic simulation platform. Methodologically, it designs a multi-agent system integrating heterogeneous households, firms, a central bank, and the government; supports both rule-based and reinforcement learning (RL) policies—specifically PPO and SAC—and pioneers the integration of the OpenAI Gym interface into macroeconomic simulation. Micro-level behavioral calibration is coupled with macro-level dynamics to enable exogenous shock modeling and counterfactual causal analysis grounded in real U.S. economic data. Contributions include: (1) the first systematic integration of deep RL into a general-purpose macroeconomic simulation framework; (2) empirical validation across two canonical scenarios—adaptive learning of employment preferences among skill-heterogeneous households, and evolutionary pricing responses of firms following a firm-specific productivity shock—demonstrating that learning agents substantially reshape equilibrium trajectories; and (3) an open-source, reproducible, and extensible simulation infrastructure.

This study addresses the ad hoc division between calibrated and estimated parameters in structural modeling, which often lacks a systematic foundation and can induce substantial bias due to calibration errors. For the first time, the partitioning problem is formalized as an optimization task, and a sensitivity-minimization criterion is proposed for selecting the optimal split. Specifically, a sensitivity statistic—constructed from local derivatives—quantifies how target estimates respond to perturbations in calibrated parameters. The method selects the partition that minimizes this statistic, thereby reducing worst-case local bias. Notably, it avoids repeated re-estimation and is applicable across a broad class of structural models. An application to a New Keynesian model demonstrates that the chosen partition significantly enhances estimation robustness and credibility under sizable calibration errors.

This study addresses the challenge that existing methods struggle to simultaneously reconcile the micro-level structure of enterprise supply chains with macroeconomic input-output tables. To bridge this gap, the authors propose an efficient synthetic network generation approach based solely on publicly available data. By integrating inter-firm connection topology with macroeconomic input-output constraints, the method achieves—without requiring proprietary information—the first scalable and reproducible synthetic supply network that maintains both microscopic realism and macroscopic consistency. The resulting networks accurately replicate key statistical properties observed in real-world data, thereby providing high-quality foundational inputs for large-scale economic modeling.

This study addresses the challenge posed by the “curse of dimensionality,” which renders conventional numerical methods ineffective for high-dimensional dynamic stochastic models in economics and finance. To overcome this limitation, the work proposes an innovative integration of deep learning and dynamic economic modeling by combining deep equilibrium networks, physics-informed neural networks, differentiable surrogate models, and Gaussian processes. Enhanced with active learning and dimensionality reduction techniques, the proposed framework efficiently solves heterogeneous-agent, macro-financial, and climate-economy models characterized by extremely large continuous state spaces. The approach not only facilitates structural estimation and policy simulation but also substantially improves computational efficiency and estimation accuracy. Broad applicability is further supported through open-source, reproducible code.

This study investigates the statistical and numerical convergence properties of stochastic equilibrium models, establishing their existence and enhancing solution accuracy. Building on SELCKE theory, the authors integrate eigenvalue analysis, higher-order perturbation expansions, and a recursive equilibrium framework to construct a simulation procedure that verifies the existence of stochastic equilibria and elucidates the geometric convergence mechanism toward long-run equilibrium. Key contributions include characterizing conditions under which faster convergence rates and super-consistent parameter estimation emerge under higher-order shocks, demonstrating equivalence between menu-cost and Calvo pricing models, proving that the stochastic steady state yields the optimal perturbation solution, deriving an upper bound for the peak timing of impulse responses, validating the empirical plausibility of Taylor contracts, and effectively resolving boundary-induced blow-up issues in objective functions.

Standard macroeconomic models struggle to account for the sluggish response of consumption to policy shocks, often resorting to exogenous frictions such as habit formation. This paper proposes an endogenous theory of macroeconomic inertia by embedding survey-based expectations on income and interest rates—instead of rational expectations—into a heterogeneous-agent general equilibrium framework that combines Blanchard’s perpetual youth structure with Bewley’s incomplete markets. An unobserved components model of expectations is incorporated to capture extrapolative biases. The approach successfully replicates consumption inertia and demonstrates that policy designs weakening expectation anchoring—such as gradualist monetary policy or delayed fiscal financing—substantially prolong transmission lags, thereby offering a novel microfoundation for macroeconomic inertia and new implications for policy design.