{"paper_id":"2507.07107","title":"Machine Learning Enhanced Multi-Factor Quantitative Trading: A Cross-Sectional Portfolio Optimization Approach with Bias Correction","authors":["Yimin Du"],"paper_date":"2025-06-02","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2507.07107v1","page_url":"https://thequant.space/flowcharts/machine-learning-enhanced-multi-factor-quantitative-trading/","scores":{"empirical_rigor":9,"math_complexity":8,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a mask‑first design for rolling‑window factor pipelines and a full open‑source quantitative trading system that includes a 213‑factor GPU engine, an Adjusted‑MSE loss, block‑bootstrap GBM augmentation, and Markowitz‑Ledoit‑Wolf portfolio optimisation.","quote":"We resolve this with a mask-first design: a Boolean tradability mask is constructed at data load time and threaded through every operator, so that no window ever reads a non-tradable price.","summary":"This paper presents a comprehensive machine learning framework for quantitative trading that achieves superior risk-adjusted returns through systematic factor engineering, real-time computation optimization, and cross-sectional portfolio construction. Our approach integrates multi-factor alpha…","strategy":"mask-first multi-factor ML trading system"},"assumptions":{"status":"available","items":[{"assumption":"Daily price‑move limits (+/-10% main‑board, +/-20% STAR/ChiNext) make some closing prices non‑executable.","type":"market","stated":"explicit","quote":"Rolling-window factor pipelines for Chinese A-share markets contain a subtle but costly flaw: daily price-move limits (+/-10% main-board, +/-20% STAR/ChiNext) render a fraction of closing prices non-executable"},{"assumption":"Only trades on days where both current and next‑day masks are 1 are used for training.","type":"data","stated":"explicit","quote":"A sample enters the dataset only if both Mt,i = 1 and Mt+1,i = 1—this bidirectional check is critical because even if factors are clean at t, if the stock hits a limit at t+1, the label return is non-executable."},{"assumption":"Synthetic returns are generated from per‑stock drift and volatility estimated on masked history and follow a log‑normal GBM with block‑bootstrap sampling.","type":"model","stated":"explicit","quote":"Generate synthetic log-returns: ˜rt,i = ˆµi + ˆσi · ϵt where innovations ϵt ∼N(0, 1) are drawn in 21-day blocks (one calendar month) to preserve short-range serial correlation."},{"assumption":"Transaction costs are modeled as a linear cost of 5–8 basis points per unit of turnover.","type":"execution","stated":"explicit","quote":"Transaction costs.. Net returns deduct a linear cost: rnet_t = w⊤_{t‑1}rt −c·∥wt −wt‑1∥1 with c = 5–8 bps per unit of turnover."},{"assumption":"The strategy is constrained to long‑only positions due to A‑share short‑selling restrictions.","type":"market","stated":"explicit","quote":"Long-only constraint.. The system is constrained to long-only due to A-share short-selling restrictions."},{"assumption":"The linear transaction‑cost model is optimistic for large portfolios where market impact is convex.","type":"execution","stated":"explicit","quote":"Our 5–8 bps linear cost model is optimistic for large portfolios where market impact is convex in trade size."}]},"datasets":{"status":"available","items":[{"name":"Synthetic panel","provider":null,"kind":"synthetic","access":"public","period":"≈14 years from 2010-01-04","universe":"3 000 stocks × 3 500 days","quote":"Synthetic panel.. 3 000 stocks × 3 500 days (≈14 years from 2010-01-04), single-factor GBM with market β = 0.55, annual drift 7 %, volatility 32 %, halt probability 1 %, ±10 % price limit."},{"name":"Real A-share panel","provider":"Tushare","kind":"market-prices","access":"commercial","period":"January 2015–December 2024","universe":"≈3 200 active stocks after masking","quote":"Real A-share panel.. Daily OHLCV from Tushare, January 2015–December 2024, average ≈3 200 active stocks after masking.","known_id":"tushare"}],"known":[],"sample":{"start":"2022","end":"2024","market":"Chinese A-share stocks","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Linear cost of 5–8 bps per unit of turnover is subtracted from net returns.","quote":"Transaction costs.. Net returns deduct a linear cost: rnet_t = w⊤_{t‑1}rt −c·∥wt −wt‑1∥1 with c = 5–8 bps per unit of turnover.","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Linear cost of 5–8 bps per unit of turnover is subtracted from net returns.","card_quote":"Transaction costs.. Net returns deduct a linear cost: rnet_t = w⊤_{t‑1}rt −c·∥wt −wt‑1∥1 with c = 5–8 bps per unit of turnover.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The mask‑first contract alone adds 0.44 Sharpe points, the largest single component contribution.","kind":"performance","quote":"Ablation shows the mask contract is the single largest contributor (+0.44), exceeding any model or loss choice."},{"finding":"Full system achieves annualised Sharpe 2.05 on synthetic data and 1.63 on real A‑share data (2022–2024).","kind":"performance","quote":"The system achieves Sharpe 2.05 on synthetic data and 1.63 on real A-share data (2022–2024), with the mask contract alone accounting for 0.44 Sharpe points."},{"finding":"Adjusted‑MSE loss contributes an additional 0.27 Sharpe points.","kind":"performance","quote":"The Adjusted‑MSE loss adds 0.27, GBM augmentation adds 0.19, and Ledoit–Wolf shrinkage adds 0.18."},{"finding":"Block‑bootstrap GBM augmentation adds 0.19 Sharpe points, especially benefiting the Transformer model.","kind":"performance","quote":"The Transformer benefits more from GBM augmentation (+0.19 Sharpe) than the MLP (+0.12)."},{"finding":"Upstream contamination (using non‑executable prices) inflates IC by 18% and reduces realised Sharpe by 0.44 points.","kind":"predictive-accuracy","quote":"On real A-share data it inflates apparent information coefficient by 18% while reducing realised Sharpe by 0.44 points, because the model learns to predict returns it cannot trade."}],"reported_performance":{"strategy":"mask-first multi-factor ML trading system","sharpe":1.63,"sharpe_benchmark":null,"annual_return_pct":20,"total_return_pct":null,"max_drawdown_pct":null},"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":25,"citations":{"cited_by_in_corpus":2,"by_independent_authors":2,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":true,"code_url":"https://github.com/initial-d/ml-quant-trading","note":null},"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":"2026-11-14T17:36:32Z"},"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/"}