{"paper_id":"2403.10273","title":"Optimal Portfolio Choice with Cross-Impact Propagators","authors":["Eduardo Abi Jaber","Eyal Neuman","Sturmius Tuschmann"],"paper_date":"2024-03-15","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2403.10273v1","page_url":"https://thequant.space/flowcharts/optimal-portfolio-choice-with-cross-impact-propagators/","scores":{"empirical_rigor":2,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a continuous‑time optimal portfolio choice model that incorporates transient cross‑impact driven by a matrix‑valued Volterra propagator together with temporary price impact, and provides explicit solutions.","quote":"We consider a class of optimal portfolio choice problems in continuous time where the agent's transactions create both transient cross-impact driven by a matrix-valued Volterra propagator, as well as temporary price impact.","summary":"We consider a class of optimal portfolio choice problems in continuous time where the agent’s transactions create both transient cross-impact driven by a matrix-valued Volterra propagator, as well as temporary price impact. We formulate this problem as the maximization of a revenue-risk functional…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Investor’s trading activity causes linear temporary impact on execution prices given by ½ Λ u_t, where Λ is a positive‑definite (not necessarily symmetric) matrix.","type":"execution","stated":"explicit","quote":"(2.4) We assume that the investor’s trading activity causes linear temporary impact on the assets’ execution prices given by 1 2Λut, 0 ≤t ≤T, (2.5) where Λ ∈RN×N is a positive definite (not necessarily symmetric) matrix, i.e. x⊤Λx > 0 for all x ∈RN.","support":"full"},{"assumption":"The price distortion created by trading is given by D_u(t)=∫_0^t G(t,s)u_s ds, where G is a matrix of Volterra kernels (G(t,s)=0 for t<s).","type":"model","stated":"explicit","quote":"The investor’s trading activity also creates a price distortion which is given by Du t := Z t 0 G(t, s)usds, 0 ≤t ≤T, (2.6) where G : [0, T]2 →RN×N is a matrix of Volterra kernels, i.e. each entry in the matrix satisfies Gij(t, s) = 0 for t < s.","support":"full"},{"assumption":"The convolution kernel H(t) used to build G(t,s)=1_{t≥s}H(t−s) is nonincreasing, convex, nonnegative‑definite and symmetric for all t∈(0,T).","type":"model","stated":"explicit","quote":"Assume that H satisfies the following assumptions: • nonincreasing, i.e. the function t 7→x⊤H(t)x is nonincreasing on (0, T) for every x ∈RN, • convex, i.e. the function t 7→x⊤H(t)x is convex on (0, T) for every x ∈RN, • nonnegative, i.e. the matrix H(t) is nonnegative definite for any t ∈(0, T), • symmetric, i.e. the matrix H(t) is symmetric for any t ∈(0, T).","support":"full"},{"assumption":"The risk term follows a Markowitz‑type quadratic form with covariance matrix Σ and risk‑aversion parameter γ≥0.","type":"market","stated":"explicit","quote":"The last term in (2.10) represents the portfolio’s risk in the sense of Markowitz [25] (see also Section 1 of [17]), where Σ was defined in (2.2) and γ ≥0 is a risk aversion parameter.","support":"full"}]},"datasets":{"status":"none_found","items":[],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"none"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R4a","candidates":{"front_matter":"n/a","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Linear temporary impact ½ Λ u_t is included in the objective; cross‑impact enters via the propagator G(t,s).","card_quote":"(2.4) We assume that the investor’s trading activity causes linear temporary impact on the assets’ execution prices given by 1 2Λut, 0 ≤t ≤T, (2.5) where Λ ∈RN×N is a positive definite (not necessarily symmetric) matrix, i.e. x⊤Λx > 0 for all x ∈RN. The investor’s trading activity also creates a price distortion which is given by Du t := Z t 0 G(t, s)usds, 0 ≤t ≤T, (2.6) where G : [0, T]2 →RN×N is a matrix of","meaning":"the paper reports no trading or execution result; the cost the extractor found is a model parameter"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"The maximization problem is solved explicitly in terms of operator resolvents by reducing the first‑order condition to a coupled system of stochastic Fredholm equations of the second kind.","kind":"theoretical","quote":"We solve the maximization problem explicitly in terms of operator resolvents, by reducing the corresponding first order condition to a coupled system of stochastic Fredholm equations of the second kind and deriving its solution.","support":"full"},{"finding":"Sufficient conditions on the matrix‑valued propagator are provided to ensure the model does not permit price manipulation.","kind":"theoretical","quote":"We then give sufficient conditions on the matrix-valued propagator so that the model does not permit price manipulation.","support":"full"},{"finding":"Numerical illustrations show that cross‑impact can reverse trading direction to exploit differing alpha decays, allowing profit from both persistent signals and cross‑impact effects.","kind":"performance","quote":"(b) In the presence of positive alpha signals on both assets, cross-impact leverages the different alpha decays. The optimal strategy may involve selling or shorting the asset with the fastest alpha decay to leverage the cross-impact effect that decreases the price of the other asset with the slowest alpha decay.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":14,"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":2,"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":2,"extracted_at":"2026-11-14T19:58:30Z"},"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/"}