{"paper_id":"2409.15459","title":"Position-building in competition with real-world constraints","authors":["Neil A. Chriss"],"paper_date":"2024-09-23","paper_type":"methodological","paper_url":"https://arxiv.org/pdf/2409.15459v2","page_url":"https://thequant.space/flowcharts/position-building-in-competition-with-real-world-constraints/","scores":{"empirical_rigor":4,"math_complexity":8,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper extends the optimal‑trading framework to a quadratic programming formulation that can incorporate real‑world constraints and computes two‑trader equilibria via an iterative algorithm.","quote":"This paper extends the optimal-trading framework developed in arXiv:2409.03586v1 to compute optimal strategies with real-world constraints. While the former paper relies on methods from the calculus of variations and optimal strategies arise as the solution of partial differential equations, the current paper demonstrates that the entire framework may be re-framed as a quadratic programming problem and cast in this","summary":"This paper extends the optimal-trading framework developed in arXiv:2409.03586v1 to compute optimal strategies with real-world constraints. The aim of the current paper, as with the previous, is to study trading in the context of multi-player non-cooperative games. While the former paper relies on…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Traders act in a non‑cooperative game; collusion is prohibited.","type":"model","stated":"explicit","quote":"The nature of the competition studied is non-cooperative, collusion is strictly prohibited.","support":"full"},{"assumption":"Trading strategies are static (time‑invariant) over the trading horizon.","type":"model","stated":"explicit","quote":"A feature and limitation of this paper and [Chriss(2024)] is that the strategies considered are static throughout the course of trading.","support":"full"},{"assumption":"All strategies start at zero inventory and end at the target quantity at t=1.","type":"execution","stated":"explicit","quote":"For this paper all trading strategies start at zero and end at the target quantity. In both cases strategies satisfy the boundary conditions of vanishing at t = 0 and acquiring λi units of stock at time t = 1.","support":"full"},{"assumption":"Aggregate trading of all participants determines the marginal price impact (linear market impact).","type":"market","stated":"explicit","quote":"Market impact: When n traders are in competition it is assumed that their aggregate trading exerts the marginal influence on the price of the stock. For strategies ai we model the cost paid by trader i at time t as proportional to:  X j ˙aj(t)  ˙ai(t) + κ  X j aj(t)  ˙ai(t) (1.1) This cost says that relative to the price at time 0 and executing every trade at the prevailing price at the time of trade depends","support":"full"},{"assumption":"The market impact coefficient κ splits temporary and permanent impact and is treated as a constant parameter.","type":"model","stated":"explicit","quote":"The parameter κ is called the market impact coefficient and represents the relative contribution between the temporary and permanent impact terms.","support":"full"},{"assumption":"Constraints such as position caps, channel limits, end‑strategy fractions, short‑selling and no‑sell can be expressed as linear inequalities on the strategy path.","type":"other","stated":"explicit","quote":"ader may be subject to restrictions on maximum position size which would lead to a constraint of the form, say, a(t) ≤1 + ρ for all t; • Channel constraints (see Section 3.3): a trader may wish to confine the trajectory of the trading strategy to a certain channel. • End-strategy constraints (see Section 3.4): a trader may wish to have a certain fraction of the target quantity acquired by no later than a certain","support":"full"}]},"datasets":{"status":"none_found","items":[],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Cost is modeled as a sum of a temporary impact term and a permanent impact term proportional to the aggregate trading rate, with market impact coefficient κ.","quote":"For strategies ai we model the cost paid by trader i at time t as proportional to:  X j ˙aj(t)  ˙ai(t) + κ  X j aj(t)  ˙ai(t) (1.1) This cost says that relative to the price at time 0 and executing every trade at the prevailing price at the time of trade depends on the aggregate quantity traded at time t by all traders and the net quantity acquired from time t = 0 to t. The parameter κ is called the market","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Cost is modeled as a sum of a temporary impact term and a permanent impact term proportional to the aggregate trading rate, with market impact coefficient κ.","card_quote":"For strategies ai we model the cost paid by trader i at time t as proportional to:  X j ˙aj(t)  ˙ai(t) + κ  X j aj(t)  ˙ai(t) (1.1) This cost says that relative to the price at time 0 and executing every trade at the prevailing price at the time of trade depends on the aggregate quantity traded at time t by all traders and the net quantity acquired from time t = 0 to t. The parameter κ is called the market","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"The algorithm reproduces the dynamic adjustment process to a two‑trader equilibrium and shows that equilibrium trading costs can be higher than the risk‑neutral (non‑equilibrium) costs.","kind":"theoretical","quote":"The algorithm allows one to understand the entire path leading to equilibrium and helps to explain certain features of equilibria such as in certain cases the cost of trading in equilibrium are higher than non-equilibrium, which can be seen as a direct consequence of the non-cooperative nature of the games. The conclusion is that in certain instances where two traders are trading approximately the same quantity, the","support":"full"},{"finding":"When collusion is prohibited, both traders incur higher total trading costs in equilibrium compared to both trading the risk‑neutral strategy.","kind":"negative","quote":"However, because the traders cannot coordinate their trading they each must successively fashion the best-response to the other’s best response and inevitably pay more. Of interest is for the two rightmost charts the total cost of trading is higher for both traders than it would be if they both traded a risk-neutral strategy as in initiation – they would both be better off, but because collusion is prohibited, they","support":"full"},{"finding":"Constrained equilibria (e.g., with over‑buying and short‑selling limits) also result in higher costs than the risk‑neutral baseline, and the larger trader’s cost rises as the smaller trader reduces her own cost.","kind":"performance","quote":"The solutions presented are the reconstructed functions a∗ and b∗ λ in Eq. (5.9), representing constrained equilibrium trading strategies, subject to constraints on overbuying and short selling, see 3.2 and 3.5. Of interest is that for the left column of charts the total cost of trading is higher for both traders than it would be if they both traded a risk-neutral strategy as in initiation – they would both be","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":17,"citations":{"cited_by_in_corpus":3,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":3},"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":"2026-11-14T19:55:05Z"},"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/"}