Selected work

Representative Papers

Measuring the Arrow of Time: Identification, Estimation, and Inference for Directional Structure in Multivariate Time Series

Aug 13, 2026

This study addresses the challenge of accurately identifying directional lead-lag relationships among variables in multivariate time series. It proposes a novel framework grounded in temporal irreversibility, rigorously defining directionality as behavioral asymmetry under time reversal. Central to this approach is the introduction of a circulation matrix derived from lagged covariances as the core measure of directionality. The methodology incorporates an unbiased estimator, cross-fitting debiasing, delete-block jackknife standard errors, and an exact randomization test under a block-wise null hypothesis, with extensions to nonlinear settings. Evaluated on four simulated systems with known directional structures, the proposed method substantially outperforms conventional approaches—including correlation networks, Granger causality, and transfer entropy—demonstrating markedly reduced misidentification of directional relationships.

0 citationsRead paper

Choice at Finite Capacity: The Bounded Agent as an Information Channel and the Recovery of Walrasian Demand

Jul 14, 2026

This study challenges the traditional economic assumption of fully rational consumers by modeling them as information processors constrained by finite channel capacity, where choices are represented as probability distributions over consumption bundles rather than deterministic selections. Integrating utility maximization with an information-theoretic framework, the paper introduces an attention-constrained information compression mechanism and derives a closed-form solution for demand responses. The model demonstrates that price effects on demand arise jointly from budget constraints and information compression: in the zero-attention limit, behavior collapses to habitual choice, while infinite attention recovers Walrasian demand. Theoretical analysis establishes symmetry of the demand-price response matrix and proves that own-price demand decreases under tight budget constraints. Full analytical solutions are obtained for the case of quadratic utility with two goods.

0 citationsRead paper

Rational Bubbles at the Spectral Edge: An Operator-Spectral Theory of Fragility, Identification and Finite-Sample Certification

Jul 04, 2026

This study addresses the challenge of identifying systemic market vulnerability and quantifying its critical threshold prior to price bubble bursts. By integrating rational bubble theory with an analysis of dominant factor strength, the authors develop a data-driven framework grounded in the co-movement structure of asset returns and discount rates. This approach uniquely unifies rational bubbles, crisis thresholds, and the spectral edge of the covariance matrix under the observable construct of a “vulnerability edge,” achieving empirical verifiability even in finite samples. Combining operator spectral theory, factor modeling, and statistical inference, the method is validated on 18 global equity indices from 2004 to 2024, revealing that dominant factors intensify markedly during crises, the number of independent factors declines from approximately six to four, and crossings of the vulnerability edge align closely with actual crisis onsets.

0 citationsRead paper

Equilibrium as a Limit: The Competitive Canon Nested in an Adaptive, Information-Theoretic Economy

Jun 26, 2026

This study addresses the lack of uniqueness, stability, and attainability of competitive equilibria in general equilibrium theory by modeling the economic system as an asymptotically mean-stationary information process composed of agents with finite information capacity. Within an information-theoretic framework, the authors integrate finite-capacity channel models, statistical dependence operators, and multi-parameter joint limit analysis to rigorously embed the classical Walrasian equilibrium—as a zero-entropy limit—within an adaptive setting for the first time. The analysis demonstrates that as the entropy rate approaches zero and channel capacity diverges, the system converges to a rational expectations competitive equilibrium. Moreover, it uncovers positive-entropy dynamic structures in non-equilibrium states that lie beyond the descriptive scope of classical theory, thereby enriching the dynamic foundations of equilibrium analysis.

0 citationsRead paper

Econstellar: An Open-Source AI-Augmented Research Engine for Computational Financial Econometrics

Jun 04, 2026

This study addresses the high computational costs and reproducibility challenges prevalent in financial econometrics research. To overcome these barriers, we propose an open-source, browser-based AI-augmented research engine that dynamically schedules computationally intensive tasks to hardware best suited for execution. The embedded AI assists solely in method selection and interpretation of results—never in data generation—ensuring users can reproduce empirical analyses using code identical to that in published papers. The system integrates established financial econometric techniques (notably return-based approaches for handling non-stationary price series), an AI-driven interpretability module, and a cloud-native reproducible architecture. It currently implements 17 validated econometric models, delivers real-time parameter estimates, provides public reproducibility endpoints, and has successfully replicated key findings from seminal studies on financial contagion.

0 citationsRead paper

Recent publications

Latest Papers

Measuring the Arrow of Time: Identification, Estimation, and Inference for Directional Structure in Multivariate Time Series

Aug 13, 2026

This study addresses the challenge of accurately identifying directional lead-lag relationships among variables in multivariate time series. It proposes a novel framework grounded in temporal irreversibility, rigorously defining directionality as behavioral asymmetry under time reversal. Central to this approach is the introduction of a circulation matrix derived from lagged covariances as the core measure of directionality. The methodology incorporates an unbiased estimator, cross-fitting debiasing, delete-block jackknife standard errors, and an exact randomization test under a block-wise null hypothesis, with extensions to nonlinear settings. Evaluated on four simulated systems with known directional structures, the proposed method substantially outperforms conventional approaches—including correlation networks, Granger causality, and transfer entropy—demonstrating markedly reduced misidentification of directional relationships.

0 citationsRead paper

Choice at Finite Capacity: The Bounded Agent as an Information Channel and the Recovery of Walrasian Demand

Jul 14, 2026

This study challenges the traditional economic assumption of fully rational consumers by modeling them as information processors constrained by finite channel capacity, where choices are represented as probability distributions over consumption bundles rather than deterministic selections. Integrating utility maximization with an information-theoretic framework, the paper introduces an attention-constrained information compression mechanism and derives a closed-form solution for demand responses. The model demonstrates that price effects on demand arise jointly from budget constraints and information compression: in the zero-attention limit, behavior collapses to habitual choice, while infinite attention recovers Walrasian demand. Theoretical analysis establishes symmetry of the demand-price response matrix and proves that own-price demand decreases under tight budget constraints. Full analytical solutions are obtained for the case of quadratic utility with two goods.

0 citationsRead paper

Rational Bubbles at the Spectral Edge: An Operator-Spectral Theory of Fragility, Identification and Finite-Sample Certification

Jul 04, 2026

This study addresses the challenge of identifying systemic market vulnerability and quantifying its critical threshold prior to price bubble bursts. By integrating rational bubble theory with an analysis of dominant factor strength, the authors develop a data-driven framework grounded in the co-movement structure of asset returns and discount rates. This approach uniquely unifies rational bubbles, crisis thresholds, and the spectral edge of the covariance matrix under the observable construct of a “vulnerability edge,” achieving empirical verifiability even in finite samples. Combining operator spectral theory, factor modeling, and statistical inference, the method is validated on 18 global equity indices from 2004 to 2024, revealing that dominant factors intensify markedly during crises, the number of independent factors declines from approximately six to four, and crossings of the vulnerability edge align closely with actual crisis onsets.

0 citationsRead paper

Equilibrium as a Limit: The Competitive Canon Nested in an Adaptive, Information-Theoretic Economy

Jun 26, 2026

This study addresses the lack of uniqueness, stability, and attainability of competitive equilibria in general equilibrium theory by modeling the economic system as an asymptotically mean-stationary information process composed of agents with finite information capacity. Within an information-theoretic framework, the authors integrate finite-capacity channel models, statistical dependence operators, and multi-parameter joint limit analysis to rigorously embed the classical Walrasian equilibrium—as a zero-entropy limit—within an adaptive setting for the first time. The analysis demonstrates that as the entropy rate approaches zero and channel capacity diverges, the system converges to a rational expectations competitive equilibrium. Moreover, it uncovers positive-entropy dynamic structures in non-equilibrium states that lie beyond the descriptive scope of classical theory, thereby enriching the dynamic foundations of equilibrium analysis.

0 citationsRead paper

Econstellar: An Open-Source AI-Augmented Research Engine for Computational Financial Econometrics

Jun 04, 2026

This study addresses the high computational costs and reproducibility challenges prevalent in financial econometrics research. To overcome these barriers, we propose an open-source, browser-based AI-augmented research engine that dynamically schedules computationally intensive tasks to hardware best suited for execution. The embedded AI assists solely in method selection and interpretation of results—never in data generation—ensuring users can reproduce empirical analyses using code identical to that in published papers. The system integrates established financial econometric techniques (notably return-based approaches for handling non-stationary price series), an AI-driven interpretability module, and a cloud-native reproducible architecture. It currently implements 17 validated econometric models, delivers real-time parameter estimates, provides public reproducibility endpoints, and has successfully replicated key findings from seminal studies on financial contagion.

0 citationsRead paper