{"paper_id":"2311.15974","title":"Adaptive Agents and Data Quality in Agent-Based Financial Markets","authors":["Colin M. Van Oort","Ethan Ratliff-Crain","Brian F. Tivnan","Safwan Wshah"],"paper_date":"2023-11-27","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2311.15974v1","page_url":"https://thequant.space/flowcharts/adaptive-agents-and-data-quality-in-agent-based-financial-ma/","scores":{"empirical_rigor":7,"math_complexity":4,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes the Agent-Based Market Microstructure Simulation (ABMMS), an agent‑based financial market model with realistic market mechanisms and adaptive agents.","quote":"We present our Agent-Based Market Microstructure Simulation (ABMMS), an Agent-Based Financial Market (ABFM) that captures much of the complexity present in the US National Market System for equities (NMS).","summary":"We present our Agent-Based Market Microstructure Simulation (ABMMS), an Agent-Based Financial Market (ABFM) that captures much of the complexity present in the US National Market System for equities (NMS). Agent-Based models are a natural choice for understanding financial markets. Financial…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The model implements only an equities market and does not include linked financial systems such as lending or options markets.","type":"market","stated":"explicit","quote":"Third, we chose to exclu-sively implement an equities market.","support":"full"},{"assumption":"The simulation focuses exclusively on a stock market with detailed implementations of market fragmentation, communication infrastructure, auction mechanisms, and adaptive agents.","type":"model","stated":"explicit","quote":"However, in this work we focus exclusively on a stock market with detailed implementations of market fragmentation, communication infrastructure, auction mechanisms, and adaptive agents.","support":"full"},{"assumption":"Response delays are modeled as propagation delays in the communication network, whereas many ABFMs ignore them.","type":"execution","stated":"explicit","quote":"These response delays are often so small that they are assumed to have minimal impact on trading outcomes, and thus are not implemented in many ABFMs.","support":"full"},{"assumption":"Exogenous limit‑price streams for ZIP agents are replaced by random limit prices drawn from a truncated normal distribution parameterized by current LULD bands.","type":"model","stated":"explicit","quote":"We replace this exogenous input with random limit prices drawn from a truncated normal distribution that is parameterized based on the current Limit Up-Limit Down (LULD) bands and is updated each time new LULD bands are issued (details in Appendix A.11.4).","support":"full"},{"assumption":"Initial LULD bands at market open are drawn from a uniform distribution between $90.00 and $110.00.","type":"model","stated":"explicit","quote":"Since ABMMS does not currently implement opening auctions, the LULD bands issued at the open are based on the closing from the “previous day”, which is drawn from a uniform distribution ranging from $90.00 to $110.00.","support":"full"},{"assumption":"Agents are initialized with random cash and share allocations.","type":"model","stated":"explicit","quote":"Each trader is randomly assigned an initial allocation of cash and shares of stock, see Table S22 for details.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"ABMMS generated simulation data","provider":null,"kind":"synthetic","access":"proprietary","period":null,"universe":null,"quote":"As a baseline, we populate ABMMS with simple trading agents and investigate properties of the generated data.","support":"full"}],"known":[{"id":"alpha-vantage","name":"Alpha Vantage","provider":"Alpha Vantage","kind":"market-prices","access":"public","quote":"We source minute resolution data from Alpha Vantage [43] for 30 US stocks: AAPL, AXP, BA, CAT, CSCO, CVX, DD, DIS, GE, GS, HD, IBM, INTC, JNJ, JPM, KO, MCD, MMM, MRK, MSFT, NKE, PFE, PG, RTX, TRV, UNH, V, VZ, WMT, and XOM.","mentions":7,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R4a","candidates":{"front_matter":"n/a","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Maker fee and taker fee define a simplified access fee schedule (traditional maker‑taker or inverted taker‑maker).","card_quote":"The primary configuration parameters for exchanges, beyond an identifier and location, are a maker fee and a taker fee. ©2023 The MITRE Corporation. Adaptive Agents and Data Quality in Agent-Based Financial Markets • 23 The maker fee and taker fee define a simplified access fee sched-ule, and allows for the implementation of a traditional maker-taker system as well as an inverted taker-maker system.","meaning":"the paper reports no trading or execution result; the cost the extractor found is a model parameter"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"The combination of detailed market mechanisms and adaptive agents produces generated data that more accurately reproduces stylized facts observed in actual markets.","kind":"statistical","quote":"The combination of detailed market mechanisms and adaptive agents leads to models whose generated data more accurately reproduce stylized facts observed in actual markets.","support":"full"},{"finding":"Baseline simulations with simple agents are compared to experimental conditions exploring market topology or meta‑reinforcement learning agents, showing the impact of these factors on data quality.","kind":"performance","quote":"We then compare the baseline with experimental conditions that explore the impacts of market topology or meta-reinforcement learning agents.","support":"full"},{"finding":"These improvements increase the utility of agent‑based financial markets as tools to inform design and policy decisions.","kind":"theoretical","quote":"These improvements increase the utility of ABFMs as tools to inform design and policy decisions.","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":15,"citations":{"cited_by_in_corpus":2,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"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":2,"extracted_at":"2026-11-14T20:00:10Z"},"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/"}