{"paper_id":"2405.07184","title":"Trade execution games in a Markovian environment","authors":["Masamitsu Ohnishi","Makoto Shimoshimizu"],"paper_date":"2024-05-12","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2405.07184v1","page_url":"https://thequant.space/flowcharts/trade-execution-games-in-a-markovian-environment/","scores":{"empirical_rigor":3,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper derives an explicit closed-form execution strategy at a Markov perfect equilibrium for two large traders.","quote":"Applying the backward induction method of dynamic programming, we provide an explicit closed-form execution strategy at a Markov perfect equilibrium.","summary":"This paper examines a trade execution game for two large traders in a generalized price impact model. We incorporate a stochastic and sequentially dependent factor that exogenously affects the market price into financial markets. Our model accounts for how strategic and environmental uncertainties…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"For all t, the parameters satisfy αt∈[0,1], βt∈[0,1], ρ∈[0,∞) and αt e^{‑ρ}+βt<1.","type":"model","stated":"explicit","quote":"For αt ∈[0, 1], βt ∈[0, 1], and ρ ∈[0, ∞), the relationship αte−ρ + βt < 1 (16) holds for all t ∈{1, . . . , T}.","support":"full"},{"assumption":"Buy‑trade and sell‑trade of a large trader induce the same instantaneous linear price impact.","type":"execution","stated":"explicit","quote":"In the rest of this paper, the buy-trade and sell-trade of a large trader are supposed to induce the same instantaneous linear price impact.","support":"full"},{"assumption":"The residual effect of past price impacts starts at zero, i.e., R0=0.","type":"model","stated":"explicit","quote":"We assume that R0 = 0, implying that there is no residual effect at the beginning of the trading window.","support":"full"},{"assumption":"Each large trader has a Constant Absolute Risk Aversion (CARA) utility function with risk‑aversion parameter γi>0.","type":"other","stated":"explicit","quote":"mics function ht (: S × Ai × Aj × (R × R) −→S) as st+1 = ht \u0000st, q1 t , q2 t , (ωt, ϵt) \u0001 , t = 1, . . . , T. (20) We assume that each large trader i ∈{1, 2} has a Constant Absolute Risk Aversion (CARA)-type Von Neumann-Morgenstern (vN-M) (or negative exponential) utility function with the absolute risk aversion parameter γi > 0.","support":"full"},{"assumption":"The price impact is transient, modeled by an exponential decay function G(t)=e^{‑ρt}.","type":"model","stated":"explicit","quote":"llowing exponential function G: R →R++: G(t) := e−ρt, (3) where ρ ∈R+ := [0, ∞) stands for the deterministic resilience speed.1 The dynamics of the residual effect of past price impacts, represented by {Rt}t∈{1,...,T}, are defined as follows: Rt+1 := t X k=1 αkλk(q1 k + q2 k)e−ρ((t+1)−k) = e−ρ t−1 X k=1 αkλk(q1 k + q2 k)e−ρ(t−k) + atλt \u0000q1 t + q2 t \u0001 e−ρ = e−ρ \u0002 Rt + αtλt(q1 t + q2 t ) \u0003 , t = 1, . . . , T, (4)","support":"full"},{"assumption":"The market price dynamics include both permanent and transient impacts and an exogenous Markovian environment.","type":"market","stated":"explicit","quote":"Therefore, the dynamics of the market price becomes Pt+1 = Pt + βtλt(q1 t + q2 t ) + It + ϵt, t = 1, . . . , T, (15) implying that we have a permanent price impact model with the effect of a Markovian environment.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Synthetic simulation data","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"In addition, our simulation-based numerical experiments suggest that the execution strategy captures various features observed in financial markets. Section 4 is devoted to simulation-based numerical experiments.","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"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":"Transient and permanent price impacts are modeled via coefficients αt, βt and λt, with exponential decay parameter ρ.","card_quote":"1] represents the linear price impact coefficient representing the temporary price impact. For αt ∈[0, 1], βt ∈[0, 1], and ρ ∈[0, ∞), the relationship αte−ρ + βt < 1 (16) holds for all t ∈{1, . . . , T}.","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 equilibrium execution strategy is generally dynamic and non‑randomized, becoming deterministic only when the Markovian environment is deterministic.","kind":"theoretical","quote":"Our theoretical results reveal that the execution strategy generally lies in a dynamic and non-randomized class; it becomes deterministic if the Markovian environment is also deterministic. If the Markovian environment is deterministic in time, so is the equilibrium execution strategy for each large trader.","support":"full"},{"finding":"Execution speed is indifferent when environmental uncertainty is large.","kind":"theoretical","quote":"The execution speed, however, seems indifferent when the environmental uncertainty (driven by the Markovian environment) is large.","support":"full"},{"finding":"The equilibrium execution strategy becomes asymmetric among traders with opposite initial inventories under the Markovian environment.","kind":"theoretical","quote":"Moreover, the execution strategy at the equilibrium becomes asymmetric among large traders with opposite initial inventories by the condition of the Markovian environment.","support":"full"},{"finding":"The strategy captures a U‑shaped total traded volume curve observed in financial markets.","kind":"theoretical","quote":"This execution strategy remarkably captures the situation that the total traded volume in a financial market seemingly forms a U-shaped trading curve.","support":"full"},{"finding":"Simulation experiments show that large traders are sensitive to environmental uncertainty, executing orders faster under low uncertainty and slower under high uncertainty.","kind":"performance","quote":"On the contrary, large traders are sensitive to environmental uncertainty, so they execute their orders under high environmental uncertainty regardless of the existence of other large traders.","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":21,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"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":"2026-11-14T19:57:25Z"},"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/"}