{"paper_id":"2309.11693","title":"Doubly Robust Mean-CVaR Portfolio","authors":["Kei Nakagawa","Masaya Abe","Seiichi Kuroki"],"paper_date":"2023-09-20","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2309.11693v1","page_url":"https://thequant.space/flowcharts/doubly-robust-mean-cvar-portfolio/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a Doubly Robust mean‑CVaR portfolio optimization method that addresses instability and incorporates uncertainty sets for the mean parameter.","quote":"Therefore in this study, we propose a Doubly Robust mean-CVaR Portfolio refined approach to the mean-CVaR portfolio optimization.","summary":"In this study, we address the challenge of portfolio optimization, a critical aspect of managing investment risks and maximizing returns. The mean-CVaR portfolio is considered a promising method due to today’s unstable financial market crises like the COVID-19 pandemic. It incorporates expected…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Φ(w, α) is a continuous function with respect to α.","type":"statistical","stated":"explicit","quote":"For simplicity, we assume that Φ(w, α) is a continuous function with respect to α."},{"assumption":"Expected returns lie in an ellipsoidal uncertainty set UE.","type":"model","stated":"explicit","quote":"Now that we can formulate the mean-multiple CVaR portfolio optimization problem, we assume the following ellipsoidal uncertainty set UE for expected returns"},{"assumption":"Asset returns are normally distributed (sub‑Gaussian).","type":"statistical","stated":"explicit","quote":"Here, note that the normal distribution N(µ, σ2) is a sub‑Gaussian distribution with parameter σ."},{"assumption":"Loss function is defined as L(w, r) := −wᵀr and asset returns have a continuous density f(r).","type":"model","stated":"explicit","quote":"Let L(w, r) be a portfolio loss function defined as L(w, r) := −wᵀr in this paper. Here, we assume that rₙ"}]},"datasets":{"status":"available","items":[{"name":"FF48","provider":"Fama and French","kind":"benchmark","access":"public","period":"January 1981 to December 2020","universe":"48 industry‑sector portfolios","quote":"We use FF48 datasets from January 1981 to December 2020 and MSCI U.S. datasets from January 2001 to December 2020 for out‑of‑sample periods."},{"name":"MSCI U.S.","provider":"MSCI","kind":"benchmark","access":"public","period":"January 2001 to December 2020","universe":"U.S. large‑ and mid‑cap stocks","quote":"The MSCI indices comprise the large and mid‑cap segments of the U.S. markets and are widely used as a benchmark for institutional investors."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"no","evidence_card":"partial","evidence_card_quoted":true},"conflict_kind":"graded","card_detail":"Turnover (TO) is used as a proxy for rebalancing costs; no explicit cost rates are incorporated in headline performance numbers.","card_quote":"Turnover (TO), a performance measure that represents the robustness of a portfolio. The TO indicates the volumes of rebalancing.","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The DR‑MCVaR portfolio exhibits lower turnover (26.85%–28.18% for FF48, 153.98%–168.67% for MSCI U.S.) than mean‑CVaR and mean‑MCVaR portfolios, indicating greater robustness.","kind":"performance","quote":"When we compare our proposed portfolio (DR-MCVaR) portfolio with others, it exhibits a notably lower TO than the mean-CVaR and mean-MCVaR portfolios. The TOs for DR‑MCVaR range between 26.85% and 28.18% for FF48 and between 153.98% and 168.67% for MSCI U.S."},{"finding":"DR‑MCVaR consistently achieves strong risk‑adjusted performance, with high returns and favorable R/R and Calmar ratios across both datasets.","kind":"performance","quote":"... consistently exhibits strong performance across most metrics in both panels, showcasing both high returns and favorable risk‑adjusted returns."},{"finding":"The proposed method can be formulated as a second‑order cone programming problem, matching the computational complexity of traditional mean‑variance optimization.","kind":"theoretical","quote":"Theoretically, the proposed method can be formulated as a second‑order cone programming problem which is the same formulation as traditional mean‑variance portfolio optimization."},{"finding":"An estimation error bound for the finite‑sample case is derived for the DR‑MCVaR method.","kind":"theoretical","quote":"In addition, we derive an estimation error bound of the proposed method for the finite‑sample case."}],"reported_performance":null,"out_of_sample":{"method":"walk-forward","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":39,"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":"2026-11-14T18:22:00Z"},"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/"}