{"paper_id":"2611.20839","title":"Stochastic Modelling of High-frequency (HF) Crypto Limit Order Book (LOB) with General Compound Marked Hawkes Process","authors":["Mine Caglar","Serhat Guldogan","Anatoliy Swishchuk"],"paper_date":"2026-11-16","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2611.20839","page_url":"https://thequant.space/flowcharts/stochastic-modelling-of-high-frequency-hf-crypto-limit-order-book-lob-with/","scores":{"empirical_rigor":7,"math_complexity":9,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a general compound marked Hawkes process (GCMHP) to model high‑frequency crypto limit order book dynamics, incorporating order‑size marks and a Markov‑chain driven price increment component.","quote":"The paper is devoted to the stochastic modelling of high-frequency crypto Limit Order Book (LOB) with general compound marked Hawkes process (GCMHP).","summary":"The paper is devoted to the stochastic modelling of high-frequency crypto Limit Order Book (LOB) with general compound marked Hawkes process (GCMHP). We consider marked Hawkes process in a general form. LLN and FCLTs with stochastic and constant centralization are stated. Numerical example for…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Condition A: finite variances Var[Xi] and Var[H(Z)] and the stationary condition \\(\\hat h = \\int_0^{\\infty} H(z)f(z)dz < 1\\).","type":"model","stated":"explicit","quote":"Throughout the paper we suppose that condition A) is satisfied: A) V ar[Xi] < +∞and V ar[H(Z)] < +∞, and we also have the following stationary condition ˆh := Z +∞ 0 H(z)f(z)dz < 1, (3) 4 where f(z) = F ′(z)-density for F(z)-cdf of Zi.","support":"full"},{"assumption":"Condition B: the mark process \\(X_i\\) is an ergodic Markov chain with stationary distribution \\(\\pi\\).","type":"model","stated":"explicit","quote":"In addition to condition A), we suppose that the following condition B) is also satis-fied: B) Markov chain Xi is ergodic with stationary distribution π = (π1, . . . , πN).","support":"full"},{"assumption":"Marks multiply the excitation kernel, so larger marks increase future intensity more than smaller marks.","type":"model","stated":"implied","quote":"Using this simple class of processes we see that the mark multiplies the exciting kernel, so that an event with a large mark value will increase the future intensity much more than an event with a small mark value would.","support":"full"},{"assumption":"Timestamps are recorded at 1‑second precision, which is considered sufficient to capture the intensity of event arrivals.","type":"data","stated":"explicit","quote":"The timestamps are recorded at a precision of 1 second, which is reasonable to accurately capture the intensity of the event arrival process.","support":"full"},{"assumption":"Log‑volume of orders follows a log‑normal distribution and is used as the mark \\(Z_i\\) in the GCMHP.","type":"data","stated":"explicit","quote":"As a result, the volume is log-normally distributed and we use log-volume sequence as the marks Zi of the GCMHP model with estimated mean µz = 13.329, and standard deviation σz = 0.452 rather than the volume itself.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"High‑frequency crypto limit order book data (BTC)","provider":"Kaggle (originally from [25])","kind":"order-book","access":"public","period":"7 April 2021 to 19 April 2021","universe":"BTC (bitcoin)","quote":"4.1 HF Crypto LOB Data The empirical analysis presented in this paper uses a high-frequency (HF) Limit Order Book (LOB) dataset for BTC (bitcoin) sourced from [24]. The data, spanning about 12 days from 7 April 2021 to 19 April 2021, capture the Level 1 activity of the order book, recording updates to the best bid and best ask prices, as well as the mid-price.","support":"full","known_id":"kaggle"}],"known":[{"id":"kaggle","name":"Kaggle","provider":"Kaggle","kind":"other","access":"public","quote":"4 Numerical Example: HF Crypto LOB In this section, we implement our results using a publicly available Kaggle data set [24], which was originally obtained from [25].","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"n/a","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"pricing-calibration","findings":[{"finding":"Law of large numbers (LLN) and functional central limit theorems (FCLTs) are derived for the GCMHP, showing convergence to a drift \\(a^*\\lambda/(1-\\hat h)\\) and a diffusion term.","kind":"theoretical","quote":"LLN and FCLTs with stochastic and constant centralization are stated. Let S(t) be a GCMHP with 0 < ˆh = R H(z)f(z)dz < 1 and a∗= P i∈X a(i) · πi.","support":"full"},{"finding":"Empirical standard deviations of the scaled price process match the theoretical diffusion coefficient, with regression \\(y = 12.8 + 7.11\\sqrt{n}\\) supporting the asymptotic approximation.","kind":"statistical","quote":"We have std{S∗ i } ≈σ s λ 1 −ˆh √n if (5) is valid. The regression curve is y = 12.8 + 7.11√n and approximation (12) clearly improves the fit of the estimated limiting diffusion coefficient σ q λ 1−ˆh.","support":"full"},{"finding":"Approximately 75% of BTC mid‑price changes are self‑generated within the market, as indicated by the estimated branching ratio.","kind":"statistical","quote":"Furthermore, the estimated branching ratio demonstrates that roughly 75% of BTC mid-price changes are self-generated within the market rather than driven by external shocks.","support":"full"},{"finding":"The empirical mean of mid‑price increments (0.0008) aligns with the predicted asymptotic drift from the GCMHP model.","kind":"statistical","quote":"Note that the empirical mean of the mid-price increments is also 0.0008 in alliance with the predicted asymptotic drift from the model.","support":"full"},{"finding":"Simulated price paths from the asymptotic SDE stay within a 95% confidence interval around the observed mid‑price series, confirming model fit.","kind":"performance","quote":"The simulated paths span approximately ±2 σ∗ around the drift and encapsulate the mid-price data within a 95% confidence interval.","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":20,"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":true,"code_url":"https://github.com/serhatgd/Crypto-Marked-Compound-Hawkes-Process","note":null},"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":3,"extracted_at":"2026-11-17T15:29:19Z"},"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/"}