{"paper_id":"2510.03129","title":"Signature-Informed Transformer for Asset Allocation","authors":["Yoontae Hwang","Stefan Zohren"],"paper_date":"2025-10-03","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2510.03129v1","page_url":"https://thequant.space/flowcharts/signature-informed-transformer-for-asset-allocation/","scores":{"empirical_rigor":6,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes the Signature‑Informed Transformer (SIT), a unified deep‑learning model that extracts path‑signature features and makes portfolio allocation decisions in a single policy.","quote":"We propose the Signature Informed Transformer to address this by unifying feature extraction and decision making into a single policy.","summary":"Robust asset allocation is a key challenge in quantitative finance, where deep-learning forecasters often fail due to objective mismatch and error amplification. We introduce the Signature-Informed Transformer (SIT), a novel framework that learns end-to-end allocation policies by directly…","strategy":"Signature-Informed Transformer (SIT)"},"assumptions":{"status":"available","items":[{"assumption":"The attention gate scalar γ is strictly positive.","type":"model","stated":"explicit","quote":"c bias matrix is scaled by a learnable, strictly positive scalar gate, γ > 0 (parameterized as γ = softplus(ˆγ))"},{"assumption":"A zero risk‑free rate is used for all risk‑adjusted metrics.","type":"market","stated":"explicit","quote":"assuming a zero risk‑free rate."},{"assumption":"Portfolios are long‑only, fully invested, and rebalanced monthly.","type":"execution","stated":"explicit","quote":"long‑only, fully invested, monthly (k‑step) rebalancing on the 40‑ and 50‑asset universes"},{"assumption":"Transaction costs are one‑way proportional fees applied per dollar traded.","type":"execution","stated":"explicit","quote":"Transaction costs are one‑way proportional fees of c ∈0, 5, 10 basis points per dollar traded."}]},"datasets":{"status":"available","items":[{"name":"S&P 100 price data","provider":"Wharton Research Data Services (WRDS)","kind":"market-prices","access":"commercial","period":"2000‑01‑01 to 2024‑12‑27","universe":"30, 40, and 50 S&P 100 companies","quote":"... The daily price data was sourced from Wharton Research Data Services (WRDS).","known_id":"wrds"},{"name":"Dow30 price data","provider":"Wharton Research Data Services (WRDS)","kind":"market-prices","access":"commercial","period":"2000‑01‑01 to 2024‑12‑27","universe":"10 and 20 assets from the DOW30","quote":"We also selected two additional portfolios of 10 and 20 assets from the DOW30 to validate performance against a different index composition","known_id":"wrds"},{"name":"CSI 300 price data","provider":"Wharton Research Data Services (WRDS)","kind":"market-prices","access":"commercial","period":"2000‑01‑01 to 2024‑12‑27","universe":"50 and 100 assets from the CSI 300 index","quote":"included two portfolios consisting of 50 and 100 assets from the CSI 300 index.","known_id":"wrds"}],"known":[{"id":"wrds","name":"WRDS","provider":"Wharton Research Data Services","kind":"other","access":"commercial","quote":"The daily price data was sourced from Wharton Research Data Services (WRDS).","mentions":2,"used_in":1}],"sample":{"start":null,"end":null,"market":"S&P 100 stocks","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":false,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"partial","evidence_card":"yes","evidence_card_quoted":true},"conflict_kind":"graded","card_detail":"One‑way proportional fees of 0, 5, or 10 basis points per dollar traded are applied in the sensitivity analysis.","card_quote":"Transaction costs are one‑way proportional fees of c ∈0, 5, 10 basis points per dollar traded.","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Replacing the CVaR objective with a risk‑neutral mean‑return objective reduces Sharpe from 0.6717 to 0.5691, showing the importance of decision‑focused loss.","kind":"performance","quote":"When the Conditional Value‑at‑Risk (CVaR) loss is replaced with a standard risk‑neutral objective (w/o CVaR), the Sharpe Ratio on the 40‑asset universe falls from 0.6717 to 0.5691."},{"finding":"Performance peaks at moderate allocation dispersion (τ ≈ 1.3) and degrades linearly with transaction costs, losing about 0.03‑0.04 Sharpe when costs increase from 0 to 10 bps.","kind":"performance","quote":"First, performance peaks at moderate dispersion, near τ ≈1.3. Second, frictions compress Sharpe roughly linearly over this range: moving from 0 to 10 bps reduces Sharpe by about 0.03–0.04 at the optimum."},{"finding":"SIT decisively outperforms traditional and advanced forecasting baselines across diverse equity universes.","kind":"performance","quote":"Experiments across diverse equity universes show our approach significantly outperforms both traditional strategies and advanced forecasting baselines."}],"reported_performance":{"strategy":"Signature-Informed Transformer (SIT)","sharpe":0.7715,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":0.3271},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":23,"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":"2026-11-14T17:25:44Z"},"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/"}