{"paper_id":"2402.10803","title":"Modelling crypto markets by multi-agent reinforcement learning","authors":["Johann Lussange","Stefano Vrizzi","Stefano Palminteri","Boris Gutkin"],"paper_date":"2024-02-16","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2402.10803v1","page_url":"https://thequant.space/flowcharts/modelling-crypto-markets-by-multi-agent-reinforcement-learni/","scores":{"empirical_rigor":8,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a multi‑agent reinforcement learning (MARL) model called SYMBA to simulate crypto markets using calibrated agents.","quote":"this study introduces a multi-agent reinforcement learning (MARL) model simulating crypto markets, which is calibrated to the Binance's daily closing prices of 153 cryptocurrencies that were continuously traded between 2018 and 2022.","summary":"Building on a previous foundation work (Lussange et al. 2020), this study introduces a multi-agent reinforcement learning (MARL) model simulating crypto markets, which is calibrated to the Binance’s daily closing prices of 153 cryptocurrencies that were continuously traded between 2018 and 2022.…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The behavior of simulated agents accurately reflects that of real‑world investors.","type":"model","stated":"explicit","quote":"i- that the behavior of the simulated agents accurately reflects that of real-world investors"},{"assumption":"Transaction limit orders in the simulated order book faithfully represent actual crypto market order dynamics.","type":"market","stated":"explicit","quote":"ii- that the transaction limit orders simulated in the order book faithfully represent the dynamics and characteristics of actual crypto market orders"},{"assumption":"Agents can only buy, sell, or hold assets (long‑only strategy).","type":"model","stated":"explicit","quote":"our approach simplifies the interaction of any agent, regardless of its behavior or strategy, into three distinct possibilities: buying, selling, or holding assets (a long-only strategy)"},{"assumption":"Network fees, risk‑free rate and APR are uniform across all simulated crypto assets.","type":"execution","stated":"explicit","quote":"we assume these financial parameters to be uniform across simulated crypto assets, with network fees set at b = 0.1%, the risk‑free rate at R = 1%, and the APR at D = 3%"},{"assumption":"Each time step corresponds to a single trading day (24/7 crypto market).","type":"data","stated":"explicit","quote":"each time step corresponds to a single trading day, or actually a trading day and night, since contrary to stock markets, crypto markets like Binance trade 24/7"},{"assumption":"The model does not include short‑selling, leveraging, derivatives, meta‑orders, market orders, intraday effects, or regulatory constraints.","type":"other","stated":"explicit","quote":"the lack of diverse trading strategies available on Binance, such as short‑selling, leveraging, derivatives, metaorders, market orders, and more. iv- Neglecting intraday and seasonal market effects. v- Disregarding legal and regulatory constraints"}]},"datasets":{"status":"available","items":[{"name":"CryptoTick daily closing prices and volumes","provider":"CryptoTick","kind":"market-prices","access":"public","period":"2018-09-27 to 2022-09-27","universe":"153 crypto assets listed on Binance","quote":"we employed high-quality, industry-standard daily closing prices and trading volumes from CryptoTick, for 153 crypto assets listed on Binance. This dataset covers the period from September 27th, 2018, to September 27th, 2022"}],"known":[{"id":"binance","name":"Binance","provider":"Binance","kind":"market-prices","access":"public","quote":"To achieve this, we employed high-quality, industry-standard daily closing prices and trading volumes from CryptoTick, for 153 crypto assets listed on Binance.","mentions":18,"used_in":4}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"n/a","evidence_card":"yes","evidence_card_quoted":true},"conflict_kind":"applicability","card_detail":"Network fees set at b = 0.1%, risk‑free rate R = 1%, APR D = 3% applied uniformly to all assets.","card_quote":"we assume these financial parameters to be uniform across simulated crypto assets, with network fees set at b = 0.1%, the risk‑free rate at R = 1%, and the APR at D = 3%","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"The calibrated SYMBA model reproduces key crypto market microstructure features such as non‑normal returns, volume clustering, and decaying price log‑return autocorrelations.","kind":"statistical","quote":"Our results offer valuable insights into the complexity and behavior of crypto markets, highlighting the model’s ability to emulate key market characteristics such as non‑normal returns, volume clustering, and decaying price log‑returns autocorrelations."},{"finding":"Simulated logarithmic price return distributions closely match those of real Binance data.","kind":"pricing-calibration","quote":"Figure 7: Comparative Distribution of Logarithmic Price Returns: The dashed black curve represents real data, while the continuous red curve represents simulated data."},{"finding":"Increasing the number of agents reduces short‑term price fluctuations linearly.","kind":"statistical","quote":"augmenting the count of agents (I) displayed a linear correlation with the reduction of short-term price fluctuations."},{"finding":"Larger gesture scalar and cointegration accuracy lead to linear increases in absolute daily price returns.","kind":"statistical","quote":"increasing the values of the gesture scalar (ζi) and the cointegration accuracy (ν) exhibited a linear escalation in absolute daily price returns."},{"finding":"Drawdown thresholds above 30% have minimal impact on agent survivability rates.","kind":"statistical","quote":"large drawdown thresholds (L > 30%) showed minimal impact on agent survivability rates."}],"reported_performance":null,"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[{"paper":{"paper_id":"2401.05393","title":"RIVCoin: an alternative, integrated, CeFi/DeFi-Vaulted Cryptocurrency","paper_date":"2023-12-19","authors":["Roberto Rivera","Guido Rocco","Massimiliano Marzo","Enrico Talin","Ammar Elsabe"],"empirical_rigor":2,"page_url":"https://thequant.space/flowcharts/rivcoin-an-alternative-integrated-cefi-defi-vaulted-cryptocurrency/"},"question":"How does increasing the number of market participants or agents affect price volatility?","this_paper_finding":"Increasing the number of agents reduces short‑term price fluctuations linearly.","other_paper_finding":"Increasing the number of market participants lowers price volatility and improves informational efficiency.","other_paper_quote":"At the same time, the volatility of the price becomes lower, as well as informational efficiency.","note":"Both findings report that more participants lower price volatility; B quantifies a linear reduction of short‑term fluctuations, while A notes a general decrease and added informational efficiency.","confidence":0.777},{"paper":{"paper_id":"2603.03671","title":"Is an investor stolen their profits by mimic investors? Investigated by an agent-based model","paper_date":"2026-03-04","authors":["Takanobu Mizuta","Isao Yagi"],"empirical_rigor":3,"page_url":"https://thequant.space/flowcharts/is-an-investor-stolen-their-profits-by-mimic-investors-investigated-by-an-agent/"},"question":"How does increasing the number of agents affect market price stability?","this_paper_finding":"Increasing the number of agents reduces short‑term price fluctuations linearly.","other_paper_finding":"Increasing the number of additional fundamental agents (AFAs) makes market prices more stable, which leads to a decrease in their profits.","other_paper_quote":"We found that in the case with increasing AFAs, market prices are made stable that leads to decrease their profits.","note":"Both findings show that more agents (linear increase) reduce short‑term price fluctuations, leading to greater stability; B notes this stability lowers profits.","confidence":0.624}],"contradicts":[{"paper":{"paper_id":"2603.03671","title":"Is an investor stolen their profits by mimic investors? Investigated by an agent-based model","paper_date":"2026-03-04","authors":["Takanobu Mizuta","Isao Yagi"],"empirical_rigor":3,"page_url":"https://thequant.space/flowcharts/is-an-investor-stolen-their-profits-by-mimic-investors-investigated-by-an-agent/"},"question":"How does increasing the number of agents affect market price stability?","this_paper_finding":"Increasing the number of agents reduces short‑term price fluctuations linearly.","other_paper_finding":"Increasing the number of additional technical agents (ATAs) makes market prices more unstable, which leads to an increase in their profits.","other_paper_quote":"In the case with increasing ATAs, market prices are made unstable that leads to gain their profits more.","note":"A finds more agents reduce short‑term price fluctuations linearly, while B finds more technical agents increase price instability and profits.","confidence":0.719},{"paper":{"paper_id":"2405.02849","title":"Modelling Opaque Bilateral Market Dynamics in Financial Trading: Insights from a Multi-Agent Simulation Study","paper_date":"2024-05-05","authors":["Alicia Vidler","Toby Walsh"],"empirical_rigor":2.5,"page_url":"https://thequant.space/flowcharts/modelling-opaque-bilateral-market-dynamics-in-financial-trad/"},"question":"How does the number of agents affect market price stability?","this_paper_finding":"Increasing the number of agents reduces short‑term price fluctuations linearly.","other_paper_finding":"Agent diversity (heterogeneous features) promotes market stability, whereas increasing the number of homogeneous agents leads to rapid instability, collapsing by the 25th time step.","other_paper_quote":"The examination of systems with identical agent and landscape feature sets, varying only in the number of agents, indicates a strong correlation between homogeneity and fragility. Systems with homogeneous agent features exhibit accelerated instability, collapsing by the 25th time step, as seen in our simulations that were designed to run until the 4000th time step.","note":"A: more agents linearly reduce short‑term price fluctuations. B: more homogeneous agents cause rapid instability, collapsing by step 25.","confidence":0.516}],"qualifies":[],"compared_with":38,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":1,"mentions":0,"unclassified":0},"independent_papers":[{"paper":{"paper_id":"2511.02136","title":"JaxMARL-HFT: GPU-Accelerated Large-Scale Multi-Agent Reinforcement Learning for High-Frequency Trading","paper_date":"2025-11-03","authors":["Valentin Mohl","Sascha Frey","Reuben Leyland","Kang Li","George Nigmatulin","Mihai Cucuringu"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/jaxmarl-hft-gpu-accelerated-large-scale-multi-agent-reinfor/"},"how":"discusses_findings","quote":"In a subsequent study, they support these findings by extending their analysis to daily cryptocurrency closing price data [CITED]."}],"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":"2026-11-14T18:14:37Z"},"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/"}