{"paper_id":"2502.17417","title":"Event-Based Limit Order Book Simulation under a Neural Hawkes Process: Application in Market-Making","authors":["Luca Lalor","Anatoliy Swishchuk"],"paper_date":"2025-02-24","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2502.17417v1","page_url":"https://thequant.space/flowcharts/event-based-limit-order-book-simulation-under-a-neural-hawke/","scores":{"empirical_rigor":7,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes an event-driven Limit Order Book simulation model based on 12 key LOB events using a Neural Hawkes process.","quote":"In this paper, we propose an event-driven Limit Order Book (LOB) model that captures twelve of the most observed LOB events in exchange-based financial markets.","summary":"In this paper, we propose an event-driven Limit Order Book (LOB) model that captures twelve of the most observed LOB events in exchange-based financial markets. To model these events, we propose using the state-of-the-art Neural Hawkes process, a more robust alternative to traditional Hawkes…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Only one LOB event can occur at any given time.","type":"model","stated":"explicit","quote":"Consider (N(t) = (N1(t), N2(t), ..., Nm(t)) to be a simple multivariate counting process, where only one event can occur at any given time. This assumption is realistic for LOB data due to its high-resolution nature, where each LOB event has a unique timestamp, almost surely."}]},"datasets":{"status":"available","items":[{"name":"LOBSTER","provider":null,"kind":"order-book","access":"public","period":null,"universe":"5 stocks (AAPL, AMZN, GOOG, INTC, MSFT)","quote":"Table 1: The total number of occurrences of each LOB event type from up to ten LOB levels in the LOBSTER dataset on June 21st, 2012, along with the probability of each event occurring based on these 5 stocks.","known_id":"lobster"},{"name":"LOBSTER sample datasets","provider":null,"kind":"order-book","access":"public","period":null,"universe":"AAPL, AMZN, GOOG, INTC, MSFT","quote":"Table 2: Training and Testing Results in AAPL, AMZN, GOOG, INTC and MSFT from the LOBSTER sample datasets.","known_id":"lobster"}],"known":[{"id":"lobster","name":"LOBSTER","provider":"LOBSTER","kind":"order-book","access":"commercial","quote":"Our model is trained using the LOBSTER data described earlier, where we split the data into 60% training, 20% validation, and 20% testing.","mentions":6,"used_in":2}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"l2-book"},"transaction_costs":{"status":"not_extracted","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"none","rule":"R0","candidates":{"front_matter":"unclear","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"the classifier was not confident"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"The simulated midprice process reproduces broad volatility and excess kurtosis dynamics of real data, though it struggles with skewness and Hurst exponent details.","kind":"statistical","quote":"Our empirical results show that many of the broad dynamics seen in the simulated data are in line with the real data, as seen by the volatility and excess kurtosis measures, while the model still struggles with some of the finer details, evidenced by the absolute skewness and Hurst exponent measures."},{"finding":"Training and testing losses decrease and accuracies increase over epochs, with good generalization between train and test sets.","kind":"predictive-accuracy","quote":"Table 2 shows us that the training results generalize to the testing results quite well in this setting."},{"finding":"The deep RL market‑making agent achieves very similar performance on simulated and real data.","kind":"performance","quote":"very similar results were achieved under the simulated and real data."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":26,"citations":{"cited_by_in_corpus":3,"by_independent_authors":3,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"unclassified":2},"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":1,"extracted_at":"2026-11-14T17:44:29Z"},"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/"}