{"paper_id":"2601.04062","title":"Smart Predict--then--Optimize Paradigm for Portfolio Optimization in Real Markets","authors":["Wang Yi","Takashi Hasuike"],"paper_date":"2026-01-07","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2601.04062v2","page_url":"https://thequant.space/flowcharts/smart-predict--then--optimize-paradigm-for-portfolio-optimiz/","scores":{"empirical_rigor":8,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a Smart Predict‑then‑Optimize (SPO) decision‑focused learning framework for portfolio optimization using linear predictors and transaction‑cost‑aware models.","quote":"the Smart Predict–then–Optimize (SPO) paradigm for portfolio optimization in real markets, which explicitly aligns the learning objective with downstream portfolio decision quality rather than pointwise prediction accuracy.","summary":"Improvements in return forecast accuracy do not always lead to proportional improvements in portfolio decision quality, especially under realistic trading frictions and constraints. This paper adopts the Smart Predict–then–Optimize (SPO) paradigm for portfolio optimization in real markets, which…","strategy":"SPO+"},"assumptions":{"status":"available","items":[{"assumption":"Realized return vector r is observable during training.","type":"data","stated":"explicit","quote":"During training, the realized return vector r is assumed to be observable, and the corresponding oracle decision is defined as w⋆= arg max w∈W r⊤w."},{"assumption":"A proportional transaction fee of 0.005 is applied to portfolio turnover.","type":"execution","stated":"explicit","quote":"In all backtesting experiments, a proportional transaction fee of 0.005 is deducted from portfolio returns based on portfolio turnover, measured by ∥wt−wt−1∥1, to account for realistic trading frictions."},{"assumption":"Linear predictors are used for interpretability and robustness.","type":"model","stated":"explicit","quote":"In this study, we investigate decision-focused learning for portfolio optimization under a realistic rolling-window backtesting framework. We adopt linear predictors for interpretability and robustness"},{"assumption":"Rolling‑window backtest with monthly rebalancing reflects realistic trading.","type":"market","stated":"explicit","quote":"We evaluate the proposed approach on U.S. ETF data (2015–2025) using a rolling-window backtest with monthly rebalancing."},{"assumption":"Transaction‑fee coefficient γ > 0 is a fixed input to the optimization.","type":"model","stated":"explicit","quote":"γ > 0 the transaction‑fee coefficient. We define the value function f(ˆr) = max w∈W ( (ˆr −r)⊤w −γ∥w −wt−1∥1 )."},{"assumption":"Feasible set W of portfolio weights is predefined.","type":"model","stated":"explicit","quote":"where W denotes the feasible set of portfolio weights."}]},"datasets":{"status":"available","items":[{"name":"U.S. ETF data","provider":null,"kind":"market-prices","access":"unknown","period":"2015/01/01 to 2025/01/01","universe":"U.S. exchange‑traded funds (ETFs)","quote":"To evaluate the proposed SPO framework for portfolio optimization, we employ daily historical data of U.S. exchange‑traded funds (ETFs) as the experimental dataset from 2015/01/01 to 2025/01/01."}],"known":[],"sample":{"start":"2016","end":"2024","market":"U.S. ETFs","asset_class":"equities","frequency":"monthly"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"A proportional transaction fee of 0.005 is applied to returns based on portfolio turnover (L1 norm).","quote":"In all backtesting experiments, a proportional transaction fee of 0.005 is deducted from portfolio returns based on portfolio turnover, measured by ∥wt−wt−1∥1, to account for realistic trading frictions.","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"A proportional transaction fee of 0.005 is applied to returns based on portfolio turnover (L1 norm).","card_quote":"In all backtesting experiments, a proportional transaction fee of 0.005 is deducted from portfolio returns based on portfolio turnover, measured by ∥wt−wt−1∥1, to account for realistic trading frictions.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Decision‑focused learning (SPO+) achieves the highest annualized return (14.05%) and best risk‑adjusted performance (Sharpe 0.785, Sortino 0.728) among all strategies.","kind":"performance","quote":"Among all strategies, SPO+ achieves the highest annualized return (14.05%) and the best risk‑adjusted performance, with a Sharpe ratio of 0.785 and a Sortino ratio of 0.728."},{"finding":"Decision‑focused approaches consistently outperform predict‑then‑optimize baselines and traditional portfolio methods on a risk‑adjusted basis.","kind":"performance","quote":"Overall, decision‑focused learning approaches consistently outperform predict–then–optimize baselines and traditional portfolio optimization methods on a"},{"finding":"SPO+ with fee remains competitive, delivering strong risk‑adjusted returns while explicitly accounting for trading costs.","kind":"performance","quote":"The SPO+ with fee variant also performs competitively, delivering strong risk‑adjusted returns while explicitly accounting for trading costs."},{"finding":"During the COVID‑19 market crisis, Robust‑SPO (ρ = 0.1) and SPO+ with Fee produce almost overlapping NAV paths, indicating robustness under extreme stress.","kind":"performance","quote":"During this interval, the Robust‑SPO (ρ = 0.1) and SPO+ with Fee strategies produce almost overlapping NAV paths, suggesting that these models arrive at highly similar portfolio decisions when facing severe market stress."},{"finding":"Softmax‑based decision‑focused models do not outperform other approaches in return or risk metrics.","kind":"negative","quote":"Softmax‑based strategies fail to outperform other approaches in either return or risk‑related metrics."}],"reported_performance":{"strategy":"SPO+","sharpe":0.785,"sharpe_benchmark":null,"annual_return_pct":14.05,"total_return_pct":null,"max_drawdown_pct":28.71},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":73,"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":"2026-11-14T17:17:08Z"},"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/"}