{"paper_id":"2509.09585","title":"Causal PDE-Control for Adaptive Portfolio Optimization under Partial Information","authors":["Alejandro Rodriguez Dominguez"],"paper_date":"2025-09-11","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2509.09585v2","page_url":"https://thequant.space/flowcharts/causal-pde-control-for-adaptive-portfolio-optimization-under/","scores":{"empirical_rigor":8,"math_complexity":9.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes Causal PDE‑Control Models (CPCMs), a framework that integrates structural causal drivers, nonlinear filtering, and forward‑backward PDE control to produce robust and transparent allocation rules under partial information.","quote":"We propose Causal PDE–Control Models (CPCMs), a framework that integrates structural causal drivers, nonlinear filtering, and forward–backward PDE control to produce robust and transparent allocation rules under partial information.","summary":"Classical portfolio models tend to degrade under structural breaks, whereas flexible machine-learning allocators often lack arbitrage consistency and interpretability. We propose Causal PDE-Control Models (CPCMs), a framework that links structural causal drivers, nonlinear filtering, and…","strategy":"Causal PDE-Control Models (CPCM)"},"assumptions":{"status":"available","items":[{"assumption":"Existence of Lipschitz driver dynamics and bounded observation model (Assumptions 3.11–3.16).","type":"model","stated":"explicit","quote":"Under Assumptions 3.11–3.16 in Section 3.1.1 on coefficients and the observation model"},{"assumption":"Conditional independence of asset returns given the driver vector (screening lenses).","type":"statistical","stated":"explicit","quote":"I(ri; rj | D) = 0, (Cov) Cov(ri, rj | D) = 0, (Event) P(Ei, Ej | Dℓ) −P(Ei | Dℓ)P(Ej | Dℓ) ≤ε"}]},"datasets":{"status":"available","items":[{"name":"Daily adjusted U.S. equity prices","provider":"Bloomberg","kind":"market-prices","access":"commercial","period":"January 2001 to December 2023","universe":"U.S. equity panel (>300 candidate drivers after screening)","quote":"... From an initial Bloomberg universe of > 1000 series, we retain approximately 350 candidates via a screen on availability, stability, and informational relevance","known_id":"bloomberg"}],"known":[],"sample":{"start":null,"end":null,"market":"U.S. equity panel","asset_class":"equities","frequency":"monthly"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":true,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"yes","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"CPCM solvers achieve higher Sharpe ratios, lower turnover, and more persistent premia than econometric and machine‑learning benchmarks.","kind":"performance","quote":"Empirically, on a U.S.equity panel with more than 300 candidate drivers, CPCM solvers achieve higher Sharpe ratios, lower turnover, and more persistent premia than econometric and machine-learning benchmarks."},{"finding":"V3 (PINN+HJB) variant attains the highest mean Sharpe and cost‑adjusted Sharpe across configurations.","kind":"performance","quote":"V3 (PINN+HJB) achieves the highest mean Sharpe and cost‑adjusted Sharpe, while V1 (linear+HJB) and V4 (PINN without HJB) perform comparably."},{"finding":"Particle filtering (PF) generally yields higher Sharpe and lower martingale‑defect than Extended Kalman Filter (EKF).","kind":"performance","quote":"PF achieves lower defect values on average, indicating tighter adherence to martingale pricing consistency."}],"reported_performance":{"strategy":"Causal PDE-Control Models (CPCM)","sharpe":0.72,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":null},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":33,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:27:25Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}