{"paper_id":"2409.13608","title":"A Krasnoselskii-Mann Proximity Algorithm for Markowitz Portfolios with Adaptive Expected Return Level","authors":["Yizun Lin","Yongxin He","Zhao-Rong Lai"],"paper_date":"2024-09-20","paper_type":"methodological","paper_url":"https://arxiv.org/pdf/2409.13608v2","page_url":"https://thequant.space/flowcharts/a-krasnoselskii-mann-proximity-algorithm-for-markowitz-portf/","scores":{"empirical_rigor":8,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a new Markowitz portfolio model with an adaptive expected return level (MPAERL) and an exact, convergent Krasnoselskii‑Mann Proximity Algorithm (KMPA) to solve it.","quote":"we propose an exact, convergent, and efficient Krasnoselskii‑Mann Proximity Algorithm (KMPA) based on the proximity operator and the Krasnoselskii‑Mann momentum technique to solve the proposed model.","summary":"Markowitz’s criterion aims to balance expected return and risk when optimizing the portfolio. The expected return level is usually fixed according to the risk appetite of an investor, then the risk is minimized at this fixed return level. However, the investor may not know which return level is…","strategy":"MPAERL"},"assumptions":{"status":"available","items":[{"assumption":"Risk‑free rate is set to zero.","type":"model","stated":"explicit","quote":"Since we do not consider risk‑free assets in this paper, we let rf = 0."},{"assumption":"Transaction costs follow a proportional model with rate ν.","type":"execution","stated":"explicit","quote":"We introduce the proportional transaction cost model"},{"assumption":"Data are monthly price‑relative sequences.","type":"data","stated":"explicit","quote":"All these data sets are monthly price relative sequences, which is a conventional frequency setting for long‑term PO."},{"assumption":"The proximity operator of g has a closed form.","type":"model","stated":"explicit","quote":"g ∈Γ0(Rm) has a closed form of its proximity operator"},{"assumption":"The gradient of f is Lipschitz continuous.","type":"model","stated":"explicit","quote":"f ∈Γ0(Rm) is differentiable with a Lipschitz continuous gradient"}]},"datasets":{"status":"available","items":[{"name":"FF25","provider":"Kenneth R. French’s Data Library","kind":"market-prices","access":"public","period":"Jul/1971 ∼May/2023","universe":"25 US portfolios formed on BE/ME and investment","quote":"... formed on BE/ME (book equity to market equity) and investment from the US market.","known_id":"french-library"},{"name":"FF25EU","provider":"Kenneth R. French’s Data Library","kind":"market-prices","access":"public","period":"Nov/1990 ∼May/2023","universe":"25 European portfolios formed on ME and prior return","quote":"FF25EU contains 25 portfolios formed on ME and prior return from the European market.","known_id":"french-library"},{"name":"FF32","provider":"Kenneth R. French’s Data Library","kind":"market-prices","access":"public","period":"Jul/1971 ∼May/2023","universe":"32 US portfolios developed by BE/ME and investment","quote":"FF32 contains 32 portfolios developed by BE/ME and investment from the US market.","known_id":"french-library"},{"name":"FF48","provider":"Kenneth R. French’s Data Library","kind":"market-prices","access":"public","period":"Jul/1971 ∼May/2023","universe":"48 US industry portfolios","quote":"FF48 contains 48 industry portfolios from the US market.","known_id":"french-library"},{"name":"FF100","provider":"Kenneth R. French’s Data Library","kind":"market-prices","access":"public","period":"Jul/1971 ∼May/2023","universe":"100 US portfolios formed on ME and BE/ME","quote":"FF100 contains 100 portfolios formed on ME and BE/ME","known_id":"french-library"},{"name":"FF100MEOP","provider":"Kenneth R. French’s Data Library","kind":"market-prices","access":"public","period":"Jul/1971 ∼May/2023","universe":"100 US portfolios formed on ME and operating profitability","quote":"FF100MEOP contains 100 portfolios formed on ME and operating profitability, all from the US market.","known_id":"french-library"}],"known":[{"id":"french-library","name":"Kenneth French Data Library","provider":"Kenneth R. French","kind":"market-prices","access":"public","quote":"We conduct extensive experiments on 6 benchmark data sets from Kenneth R. French’s Data Library1 (a standard and widely-used data library for long-term PO), named FF25, FF25EU, FF32, FF48, FF100 and FF100MEOP.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":"French equity data sets","asset_class":"equities","frequency":"monthly"},"data_resolution":"weekly-monthly"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Proportional transaction cost model with rate ν ranging from 0 to 0.5%, applied to portfolio rebalancing each month.","quote":"We introduce the proportional transaction cost model","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Proportional transaction cost model with rate ν ranging from 0 to 0.5%, applied to portfolio rebalancing each month.","card_quote":"We introduce the proportional transaction cost model","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"MPAERL achieves positive α factors on all six benchmark data sets, with p‑values below 0.02, indicating statistically significant outperformance.","kind":"performance","quote":"its p‑values are all smaller than 0.02, which indicates that its α factors are greater than 0 at a confidence level of 98% on all 6 data sets."},{"finding":"MPAERL attains the highest Sharpe Ratios among all methods on all data sets, surpassing both trivial and non‑trivial competitors.","kind":"performance","quote":"our MPAERL outperforms both trivial and nontrivial strategies on all data sets. MPAERL not only allows for an adaptive expected return level but also reduces the risk at this level..."},{"finding":"In maximum drawdown experiments, MPAERL shows competitive downside‑risk control, with drawdown values comparable to or lower than other methods.","kind":"performance","quote":"MPAERL also has a competitive capability of downside risk control according to the maximum drawdown experiments, which indicates that its adaptive scheme can effectively balance return and risk."},{"finding":"Across transaction‑cost rates from 0% to 0.5%, MPAERL consistently yields higher cumulative wealth than all state‑of‑the‑art competitors on all six data sets.","kind":"performance","quote":"It shows that MPAERL outperforms other state‑of‑the‑art competitors on all 6 data sets in all the cases, which suggests that MPAERL is also effective in controlling transaction costs while managing the portfolio adaptively."}],"reported_performance":{"strategy":"MPAERL","sharpe":null,"sharpe_benchmark":0.2287,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":50.12},"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"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":"2026-11-14T18:02:10Z"},"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/"}