{"paper_id":"2505.15296","title":"Agent-based Liquidity Risk Modelling for Financial Markets","authors":["Perukrishnen Vytelingum","Rory Baggott","Namid Stillman","Jianfei Zhang","Dingqiu Zhu","Tao Chen","Justin Lyon"],"paper_date":"2025-05-21","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2505.15296v1","page_url":"https://thequant.space/flowcharts/agent-based-liquidity-risk-modelling-for-financial-markets/","scores":{"empirical_rigor":5,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a novel agent‑based model that simulates transaction costs and liquidity risk for buying or selling assets, including meta‑order execution.","quote":"In this paper, we describe a novel agent-based approach for modelling the transaction cost of buying or selling an asset in financial markets, e.g., to liquidate a large position as a result of a margin call to meet financial obligations.","summary":"In this paper, we describe a novel agent-based approach for modelling the transaction cost of buying or selling an asset in financial markets, e.g., to liquidate a large position as a result of a margin call to meet financial obligations. The simple act of buying or selling in the market causes a…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Traders assume orderflow is informed and update their belief of the fundamental value whenever a trade occurs.","type":"model","stated":"explicit","quote":"we propose a mechanism where traders assume orderflow is informed and update their belief of the fundamental value as new information is revealed whenever a trade occurs."},{"assumption":"The market can be represented with minimally complex behaviours (value, trend, noise trading) that are sufficient to capture stylised market behaviours.","type":"model","stated":"explicit","quote":"Across the different models described in the literature, we observe that value, trend and noise trading strategies are sufficient to capture the stylised behaviours of the market"},{"assumption":"Limit order arrivals follow a Poisson process and market order decisions are random (Zero‑Intelligence traders).","type":"model","stated":"explicit","quote":"For example, the arrival of limit orders can be modelled as a Poisson process with each Zero‑Intelligence trader using a Bernoulli variable with parameter α_{zi} to determine whether it will submit or not a limit order at each step, as are the market orders with a parameter μ_{zi}."},{"assumption":"The ABM is calibrated to historical tick data, assuming that such data accurately reflect market microstructure.","type":"data","stated":"explicit","quote":"In this paper, we introduce a novel agent-based high-frequency market simulation that is calibrated to historical tick data ..."},{"assumption":"Transaction cost can be expressed as a function of market impact (square‑root law) plus spread cost.","type":"model","stated":"explicit","quote":"Most transaction cost model are simple, assuming a multiplicative power‑law relationship of transaction cost and volume to model the cost from market impact and a spread cost (usually defined as crossing the spread from the mid‑price)."}]},"datasets":{"status":"available","items":[{"name":"Hang‑Seng Futures Index data","provider":"Hong Kong Exchanges and Clearing Limited","kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"We demonstrate its practical application in the real world by calculating the liquidity risk for the Hang‑Seng Futures Index."},{"name":"Historical tick data","provider":null,"kind":"order-book","access":"unknown","period":null,"universe":null,"quote":"In this paper, we introduce a novel agent-based high-frequency market simulation that is calibrated to historical tick data ..."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"tick"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Costs and uncertainties are calculated via Monte‑Carlo simulations; results show a concave (square‑root) increase with order size and are compared to the Bloomberg Transaction Cost model.","quote":"ABM model calculates the costs and uncertainties of buying and selling in a market by running Monte‑Carlo simulations","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Costs and uncertainties are calculated via Monte‑Carlo simulations; results show a concave (square‑root) increase with order size and are compared to the Bloomberg Transaction Cost model.","card_quote":"ABM model calculates the costs and uncertainties of buying and selling in a market by running Monte‑Carlo simulations","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"execution-cost","findings":[{"finding":"The ABM reproduces the square‑root law: costs increase concavely with order size.","kind":"theoretical","quote":"We also observe in Figure 7(e) a concave increase in costs as we increase"},{"finding":"The ABM’s transaction‑cost estimates are of similar magnitude to the Bloomberg Transaction Cost model after calibration.","kind":"performance","quote":"we are satisfied our ABM model produces transaction costs similar in magnitude to the commonly used approximation of the Bloomberg Transaction Cost model."},{"finding":"Optimal execution results match the Almgren‑Chriss framework: front‑loaded strategy has highest cost and lowest uncertainty, balanced strategy near the minimum.","kind":"performance","quote":"Given different strategies to execute a meta‑order of 3000 over five days, our ABM model is able to reproduce similar results as the Almgren‑Chriss framework – see Figure 8"},{"finding":"The model generates realistic transient and permanent market impact that relaxes to a permanent level, consistent with practitioner expectations.","kind":"theoretical","quote":"The result of this novel approach is realistic transient and permanent market impact that emerge, that are in‑line with what practitioners would expect from the market, e.g., follows the square‑root law and relaxes to a permanent level."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":36,"citations":{"cited_by_in_corpus":1,"by_independent_authors":0,"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":1,"extracted_at":"2026-11-14T17:37:32Z"},"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/"}