{"paper_id":"2503.04072","title":"Wasserstein Robust Market Making via Entropy Regularization","authors":["Zhou Fang","Arie Israel"],"paper_date":"2025-03-06","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2503.04072v1","page_url":"https://thequant.space/flowcharts/wasserstein-robust-market-making-via-entropy-regularization/","scores":{"empirical_rigor":3,"math_complexity":8.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a robust market making framework based on Wasserstein distance that uses a stochastic policy approach enhanced by entropy regularization and shows that the robust problem can be reformulated as a convex optimization.","quote":"In this paper, we introduce a robust market making framework based on Wasserstein distance, utilizing a stochastic policy approach enhanced by entropy regularization. We demonstrate that, under mild assumptions, the robust market making problem can be reformulated as a convex optimization question.","summary":"In this paper, we introduce a robust market making framework based on Wasserstein distance, utilizing a stochastic policy approach enhanced by entropy regularization. We demonstrate that, under mild assumptions, the robust market making problem can be reformulated as a convex optimization question.…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The ask spread ϵ+ and bid spread ϵ− are strictly positive.","type":"market","stated":"explicit","quote":"Here, we assume ϵ+ > 0 and ϵ−> 0.","support":"full"},{"assumption":"The random variables ΔN⁺(ϵ⁺) and ΔN⁻(ϵ⁻) follow the probability distributions Q⁺(ϵ⁺) and Q⁻(ϵ⁻), respectively.","type":"model","stated":"explicit","quote":"We assume the random variables ∆N +(ϵ+) and ∆N −(ϵ−) are distributed according to probability dis-tributions Q+(ϵ+) and Q−(ϵ−) on R, respectively.","support":"full"},{"assumption":"The asset price change ΔS is independent of the order‑arrival measures Q⁺(ϵ⁺) and Q⁻(ϵ⁻).","type":"statistical","stated":"explicit","quote":"Here, ∆S follows a probability measure that is independent of the measures of Q+(ϵ+) and Q−(ϵ−).","support":"full"},{"assumption":"The expected change in the asset price is zero, so expected cash change equals expected wealth change.","type":"statistical","stated":"explicit","quote":"If we assume that the expected change in the asset price is zero, then it follows that the expected change in cash is equivalent to the expected change in total wealth.","support":"full"},{"assumption":"In the classical setting the market‑making policy is deterministic, mapping the state (S,Q) to fixed spreads ϵ⁺(S,Q) and ϵ⁻(S,Q).","type":"model","stated":"explicit","quote":"In the above classical setting, the policy is deterministic, meaning that given the current state (S, Q), the policy prescribes a deterministic bid and ask spread ϵ+(S, Q) and ϵ−(S, Q), respectively.","support":"full"},{"assumption":"Deterministic policies limit the choice of policies and create difficulties when converting the problem to a robust optimization setting.","type":"model","stated":"explicit","quote":"However, assuming a deterministic policy not only limits our choice of possible policies but also poses difficulties when modifying the market-making problem into a robust optimization setting.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"standardized buy and sell order data","provider":null,"kind":"order-book","access":"unknown","period":null,"universe":null,"quote":"The empirical data for standardized buy and sell orders are given by {∆e N + i }n i=1 and {∆e N − i }n i=1, respectively.","support":"full"}],"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":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"The robust market‑making optimization problem can be reformulated as a convex optimization problem under mild assumptions.","kind":"theoretical","quote":"We demonstrate that, under mild assumptions, the robust market making problem can be reformulated as a convex optimization question.","support":"full"},{"finding":"The optimal stochastic policy is given in closed form by π* (ϵ⁺,ϵ⁻) ∝ exp( … ) as stated in Theorem 1.","kind":"theoretical","quote":"The optimal policy for the problem stated in (12) is given by: π∗(ϵ+ 0 , ϵ− 0 ) = M ∗(ϵ+ 0 , ϵ− 0 ) R (R+)2 M ∗(ϵ+, ϵ−) dϵ+ dϵ−, M ∗(ϵ+, ϵ−) = exp ( 1 γ \" (A −2ηCh+(ϵ+))α∗,+ −(B −2ηCh−(ϵ−))α∗,− −η \u0010 h+(ϵ+)2β∗,+ −2h+(ϵ+)h−(ϵ−)α∗,+α∗,−+ h−(ϵ−)2β∗,−\u0011#) L(ϵ+, ϵ−), L(ϵ+, ϵ−) = exp ( 1 γ h (S + ϵ+)f +(ϵ+) −(S −ϵ−)f −(ϵ−) −ηC2i) .","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":7,"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":2,"extracted_at":"2026-11-14T19:51:53Z"},"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/"}