{"paper":{"title":"Deep Learning for Solving and Estimating Dynamic Models in Economics and Finance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Deep learning methods solve and estimate high-dimensional dynamic stochastic models in economics and finance by embedding equilibrium conditions into neural-network training.","cross_cats":["q-fin.EC"],"primary_cat":"econ.GN","authors_text":"Simon Scheidegger","submitted_at":"2026-05-14T07:33:36Z","abstract_excerpt":"This script offers an implementation-oriented introduction to deep learning methods for solving and estimating high-dimensional dynamic stochastic models in economics and finance. Its starting point is the curse of dimensionality: heterogeneous-agent economies, overlapping-generations models with aggregate risk, continuous-time models with occasionally binding constraints, climate-economy models, and macro-finance environments with many assets and frictions generate state and parameter spaces that strain classical tensor-product grid methods. The exposition is organized around four complementa"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Deep learning methods such as Deep Equilibrium Nets, Physics-Informed Neural Networks, deep surrogate models, and Gaussian-process dynamic programming can solve and estimate high-dimensional dynamic stochastic models in economics and finance that strain classical tensor-product grid methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the neural-network approximations remain accurate and stable when applied to the equilibrium conditions and dynamics of the high-dimensional models described, without introducing material bias or convergence failures.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"The paper surveys deep learning methods such as Deep Equilibrium Nets and Physics-Informed Neural Networks for solving and estimating high-dimensional dynamic stochastic models in economics and finance.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Deep learning methods solve and estimate high-dimensional dynamic stochastic models in economics and finance by embedding equilibrium conditions into neural-network training.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d8ac337d69194510bef9ebfb674dd320d82098688a67e45d261dd514c651130b"},"source":{"id":"2605.14493","kind":"arxiv","version":1},"verdict":{"id":"9284e8e1-bf7f-487e-b6df-9ae7b3e2f675","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T01:25:56.737540Z","strongest_claim":"Deep learning methods such as Deep Equilibrium Nets, Physics-Informed Neural Networks, deep surrogate models, and Gaussian-process dynamic programming can solve and estimate high-dimensional dynamic stochastic models in economics and finance that strain classical tensor-product grid methods.","one_line_summary":"The paper surveys deep learning methods such as Deep Equilibrium Nets and Physics-Informed Neural Networks for solving and estimating high-dimensional dynamic stochastic models in economics and finance.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the neural-network approximations remain accurate and stable when applied to the equilibrium conditions and dynamics of the high-dimensional models described, without introducing material bias or convergence failures.","pith_extraction_headline":"Deep learning methods solve and estimate high-dimensional dynamic stochastic models in economics and finance by embedding equilibrium conditions into neural-network training."},"references":{"count":13,"sample":[{"doi":"","year":2022,"title":"Achdou, Y., Han, J., Lasry, J.-M., Lions, P.-L., and Moll, B. (2022). Income and wealth distribution in macroeconomics: A continuous-time approach.The Review of Economic Studies, 89(1):45–86. Adjemian","work_id":"6ac84b6f-7737-4fe9-b25e-435257aed110","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2001,"title":"Solving Nonlinear and High-Dimensional Partial Differential Equations via Deep Learning","work_id":"f91727f1-bff1-4709-96f7-557ba3710616","ref_index":2,"cited_arxiv_id":"1811.08782","is_internal_anchor":true},{"doi":"","year":2020,"title":"Bayer, C. and Luetticke, R. (2020). Solving discrete time heterogeneous agent models with aggre- gate risk and many idiosyncratic states by perturbation.Quantitative Economics, 11(4):1253–","work_id":"ac9d2aba-6d6e-4e85-88bd-fbcd775a1e4d","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Belkin, M., Hsu, D., Ma, S., and Mandal, S. (2019). Reconciling modern machine-learning practice and the classical bias–variance trade-off.Proceedings of the National Academy of Sciences, 116(32):1584","work_id":"e19348b6-f797-41b7-99c2-8080f5099085","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Global solu- tions to master equations for continuous time heteroge- neous agent macroeconomic models.arXiv:2406.13726","work_id":"e3178423-b918-43f7-a215-7d726202f917","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":13,"snapshot_sha256":"18137454ecdd603c3acd6fd39aadfcd501a4a5dea0c3dccb3bc00b38be048320","internal_anchors":4},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}