Newest papers

LOB-Impact Benchmark: Emergent Market Impact in Generative Limit Order Book Models

George Nigmatulin, Valentin Mohl, Sascha Frey, Kang Li, Aramis Marti-Shahandeh, Mihai Cucuringu, Anisoara Calinescu, Jakob Foerster, Stefan Zohren · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.27100 · cited by 0 works (OpenAlex) · open-access PDF

Korean

관찰자는이 메커니즘을 잘못 평가한다면, 에이전트는 시뮬레이터에서 최적의 정책을 배울 수 있지만 실제 시장에서 잘못됩니다. 기존 메타계는 역사적 데이터와의 분포 비교를 통해 생성된 메타계를 순전히 평가하여 거래자 자신의 주문에 대한 반사적 응답을 대체로 탐구합니다. 여기서 우리는 시뮬레이터가 시뮬레이터에서 최적화 된 메타계가 실시간 시장에서 최적화 된 정책을 배울 수 있습니다. 기존 메타계는 역사적 데이터와의 분포 비교를 통해 생성된 메시지 흐름을 순전히 평가하며, 에이전트의 자신의 주문에 대한 반사적 응답은 대체적으로 탐구되지 않습니다. 여기서 우리는 미래의 물리학이 미리 알려진 기본 라인에서 검증 된

Machine translation (Meta M2M100); the English original follows.

Abstract

Realistic market impact is essential for generative limit order book (LOB) models to serve as world models for trading agents. If a simulator misprices this mechanism, agents learn policies that are optimal in the simulator but wrong in the real market. Existing metrics assess generated streams purely through distributional comparison with historical data, leaving counterfactual response to an agent's own orders largely unexplored. Here, we show that market impact emerges naturally in neural LOB generators trained solely for next-token prediction. We inject execution-style metaorders into the generated message flow of five autoregressive models across five NASDAQ stocks and measure the response using estimators from the empirical impact literature, validated on baselines whose physics is known in advance. Our benchmark evaluates any message-level generator in three phases: free generation (reading order-flow persistence against price diffusivity); metaorder injection (yielding the impact law's exponent and amplitude); and a post-execution window (measuring relaxation). In addition, comparing each visible injection to a blind, simulator-only control isolates the model's learned response from standard book mechanics. For every model on every stock, we measure how the impact responds as a function of cumulative metaorder participation. Interestingly, we observe emergent impact in generative models of market data, with exponents in the range of 0.63 to 1.21, slightly above the traditional 0.5 of the square-root law of impact. In the family of models tested, we are so far unable to observe impact decay during relaxation. We tentatively attribute this to the still-short context windows of the models. We release the benchmark as open source to validate future generative models for market data.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Counterfactual thinking, Benchmark (surveying), Context (archaeology), Market impact, Econometrics, Generative grammar

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange