Starting from the observation that econometric models have shown significantly degraded forecasting accuracy during structural breaks and crisis regimes, when past behavioural patterns can no longer serve as reliable guides for the future due to contextual changes (deglobalization, technological innovation, and reduced international cooperation), this paper examines models that have attempted to utilise nonlinear mathematics (such as that used by physicists), the related increasingly copious statistics and sentiments information that are accumulating, and Artificial Intelligence progresses. These models do not require a priori specification of structural economic equations as mandatory starting points and benefit from increasing computational power that machine learning techniques make better use of. In this framework, economic theory is not abandoned but repositioned: it enters as optional constraints, informed priors, or evaluation benchmarks rather than as the foundational architecture of the model. The paper provides a concise overview of the progress made in this domain and the results of initial experiments demonstrating the validity of AI methods in enhancing econometric forecasting.
References
[1]
Ando, A., & Modigliani, F. (1963). The Life Cycle Hypothesis of Saving: Aggregate Implications and Tests. TheAmericanEconomicReview,53, 55-84.
[2]
Biondo, A. E., & Delli Gatti, D. (2024). Introduction to the Special Issue on Two Editions of the Workshop on Economic Science with Heterogeneous Interacting Agents, Milan, 2021 (WEHIA 2021), Catania, 2022. JournalofEconomicInteractionandCoordination,19, 579-585. https://doi.org/10.1007/s11403-024-00427-x
[3]
Black, F., & Scholes, M. (1973). The Pricing of Options and Corporate Liabilities. JournalofPoliticalEconomy,81, 637-654. https://doi.org/10.1086/260062
[4]
Bollerslev, T. (1986). Generalized Autoregressive Conditional Heteroskedasticity. JournalofEconometrics,31, 307-327. https://doi.org/10.1016/0304-4076(86)90063-1
[5]
Bouchaud, J. P., & Potters, M. (2003). TheoryofFinancialRiskandDerivativePricing: From Statistical Physics to Risk Management (2nd ed.). Cambridge University Press. https://doi.org/10.1017/cbo9780511753893
[6]
Brunton, S. L., Proctor, J. L., & Kutz, J. N. (2016). Discovering Governing Equations from Data by Sparse Identification of Nonlinear Dynamical Systems. ProceedingsoftheNationalAcademyofSciences,113, 3932-3937. https://doi.org/10.1073/pnas.1517384113
[7]
Buehler, H., Gonon, L., Teichmann, J., & Wood, B. (2019). Deep Hedging. QuantitativeFinance,19, 1271-1291. https://doi.org/10.1080/14697688.2019.1571683
[8]
de Paula, A. (2017). Econometric Models of Network Formation. University College Lon-don, Cemmap and IFS.
[9]
Dvorak, T. (2023). Whitepaper:UsingAlternativeData. Oxford Economics.
[10]
Engle, R. F. (1982). Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation. Econometrica,50, 987-1007. https://doi.org/10.2307/1912773
[11]
Farmer, J. D., & Axtell, R. L. (2025). Agent-Based Modeling in Economics and Finance: Past, Present, and Future. JournalofEconomicLiterature,63, 197-287. https://doi.org/10.1257/jel.20221319
[12]
Fehr, E., & Camerer, C. F. (2007). Social Neuroeconomics: The Neural Circuitry of Social Preferences. TrendsinCognitiveSciences,11, 419-427. https://doi.org/10.1016/j.tics.2007.09.002
[13]
Frisch, R. (1933). Propagation Problems and Impulse Problems in Dynamic Economics. In EconomicEssaysinHonourofGustavCassel (pp. 171-173). Allen & Unwin.
[14]
Galilei, G. (1623). Il Saggiatore. Giacomo Mascardi.
[15]
Giannone, D., Reichlin, L., & Small, D. (2008). Nowcasting: The Real-Time Informational Content of Macroeconomic Data. JournalofMonetaryEconomics,55, 665-676. https://doi.org/10.1016/j.jmoneco.2008.05.010
[16]
Gogas, P., Papadimitriou, T., Goumenidis, P., Kontos, A., & Giannakis, N. (2025). Identification of Investment-Ready SMEs: A Machine Learning Framework to Enhance Equity Access and Economic Growth. Forecasting,7, Article No. 51. https://doi.org/10.3390/forecast7030051
[17]
Greene, W. H. (2012). EconometricAnalysis (7th ed.). Pearson Education.
[18]
Hamilton, J. D. (2006). Computing Power and the Power of Econometrics. In 50th AnniversaryoftheEconometricInstituteatErasmusUniversity (pp. 6-8, 32-38). University of California.
[19]
Hochreiter, S., & Schmidhuber, J. (1997). Long Short-Term Memory. NeuralComputation,9, 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
[20]
Italian International Economic Center (2025). https://www.iiec.world/
[21]
Jochmans, K. (2018). Semiparametric Analysis of Network Formation. JournalofBusiness&EconomicStatistics,36, 705-713. https://doi.org/10.1080/07350015.2017.1286242
[22]
Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., & Amodei, D. (2020). Scaling Laws for Neural Language Models.
[23]
Keynes, J. M. (1936). TheGeneralTheoryofEmployment,InterestandMoney. Macmillan Cambridge University Press, for Royal Economic Society in 1936.
[24]
Lim, B., Arık, S. Ö., Loeff, N., & Pfister, T. (2021). Temporal Fusion Transformers for Interpretable Multi-Horizon Time Series Forecasting. InternationalJournalofForecasting,37, 1748-1764. https://doi.org/10.1016/j.ijforecast.2021.03.012
[25]
Litterman, R. B. (1986). Forecasting with Bayesian Vector Autoregressions—Five Years of Experience. JournalofBusiness&EconomicStatistics,4, 25-38. https://doi.org/10.1080/07350015.1986.10509491
[26]
Mantegna, R. N., & Stanley, H. E. (2000). IntroductiontoEconophysics: Correlations and Complexity in Finance. Cambridge University Press. https://doi.org/10.1017/cbo9780511755767
[27]
Modigliani, F. (1944). Liquidity Preference and the Theory of Interest and Money. Econometrica,12, 45-88. https://doi.org/10.2307/1905567
[28]
Mullainathan, S., & Spiess, J. (2017). Machine Learning: An Applied Econometric Approach. JournalofEconomicPerspectives,31, 87-106. https://doi.org/10.1257/jep.31.2.87
[29]
Ortiz, A., & Rodrigo, T. (2025). Geopolitics,GeoeconomicsandRisk:AMachineLearningApproach. Preprints and Early-Stage Research.
[30]
Pateras, J., Rana, P., & Ghosh, P. (2023). A Taxonomic Survey of Physics-Informed Machine Learning. AppliedSciences,13, Article No. 6892. https://doi.org/10.3390/app13126892
[31]
Raissi, M., Perdikaris, P., & Karniadakis, G. E. (2019). Physics-Informed Neural Networks: A Deep Learning Framework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations. JournalofComputationalPhysics,378, 686-707. https://doi.org/10.1016/j.jcp.2018.10.045
[32]
Rangel, A., Camerer, C., & Montague, P. R. (2008). A Framework for Studying the Neurobiology of Value-Based Decision Making. NatureReviewsNeuroscience,9, 545-556. https://doi.org/10.1038/nrn2357
[33]
Simeone, A., Simeone, L., & Teza, G. (2025). L’AcceleratoreAI:Comeintegrareleprevisionieconometriche(Presentation). Luiss University.
[34]
Sims, C. A. (1980). Macroeconomics and Reality. Econometrica,48, 1-48. https://doi.org/10.2307/1912017
[35]
Stiglitz, J. E. (2018). Where Modern Macroeconomics Went Wrong. OxfordReviewofEconomicPolicy,34, 70-106.
[36]
Tinbergen, J. (1939). StatisticalTestingofBusinessCycleTheories,Vol.II:BusinessCyclesintheUnitedStatesofAmerica,1919-1932. League of Nations.
[37]
Wheeler, A., & Varner, J. D. (2024). ScalableAgent-BasedModelingforComplexFinancialMarketSimulations.
[38]
Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2020). A Comprehensive Survey on Graph Neural Networks. IEEETransactionsonNeuralNetworksandLearningSystems,32, 4-24. https://doi.org/10.1109/tnnls.2020.2978386
