{"paper_id":"2508.17837","title":"Bimodal Dynamics of the Artificial Limit Order Book Stock Exchange with Autonomous Traders","authors":["Matej Steinbacher","Mitja Steinbacher","Matjaz Steinbacher"],"paper_date":"2025-08-25","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2508.17837v1","page_url":"https://thequant.space/flowcharts/bimodal-dynamics-of-the-artificial-limit-order-book-stock-ex/","scores":{"empirical_rigor":2.5,"math_complexity":7.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper shows that the artificial stock market exchange (ASME) exhibits bistability with two qualitatively distinct long‑run equilibria—a degenerate zero‑price state and a persistent positive‑price equilibrium—emerging from identical initial conditions.","quote":"Specifically, we show that the ASME exhibits bistability, whereby qualitatively distinct long-run equi-libria, a degenerate state at P ∗= 0 and a persistent positive price equilibrium P ∗> 0, can emerge from identical initial conditions.","summary":"This paper explores the bifurcative dynamics of an artificial stock market exchange (ASME) with endogenous, myopic traders interacting through a limit order book (LOB). We showed that agent-based price dynamics possess intrinsic bistability, which is not a result of randomness but an emergent…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Traders are endogenous, myopic agents who place bid and ask orders in a limit order book.","type":"model","stated":"explicit","quote":"This paper explores the bifurcative dynamics of an artificial stock market exchange (ASME) with endogenous, myopic traders interacting through a limit order book (LOB).","support":"full"},{"assumption":"The market operates as a limit order book (LOB) microstructure.","type":"market","stated":"explicit","quote":"2 Related Works Our conceptualization of the artificial stock market exchange (ASME) is rooted in the fundamental functionality of a limit order book (LOB) market, a mar-ket microstructure that has evolved significantly with technological advance-ments Garman (1976); Madhavan (2000); O’Hara (2002); O’Hara (2015); Has-brouck (2007).","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Synthetic ASME simulation (50 independent repetitions)","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"The initial monetary holdings M0 = 174.4 4.2 Simulation Results Simulation results of 50 independent repetitions all starting from the same initial conditions are shown in Figure 1. Figure 1: Simulation Results Despite the same initial conditions, the trajectories fan out into different terminal states, with 22 of them settling at higher values PT > 0 and remaining 28 trajectories dropping towards PT = 0.","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"Despite identical initial conditions, simulation trajectories diverge: 22 trajectories settle at a positive terminal price (P_T > 0) and 28 trajectories converge to zero (P_T = 0).","kind":"statistical","quote":"Figure 1: Simulation Results Despite the same initial conditions, the trajectories fan out into different terminal states, with 22 of them settling at higher values PT > 0 and remaining 28 trajectories dropping towards PT = 0.","support":"full"},{"finding":"Augmented Dickey‑Fuller tests for the P* > 0 group yield low p‑values (p < 0.05), indicating stationarity of those trajectories in the weak sense.","kind":"statistical","quote":"For all statistical measures of the trajectories, the low p-values (p < 0.05) lead to the rejection of the null hypothesis, providing strong evidence that the trajectories of the P ∗> 0 group exhibit stationarity in the weak sense, meaning their first two moments are time-invariant (E[yt] = µ, Cov(yt, yt−k) = γ(k)).","support":"full"},{"finding":"Gradient Boosting Machine (GBM) predicts reaching zero with higher overall accuracy (0.89) than Logistic Regression (0.79), and shows superior precision and recall for both classes.","kind":"predictive-accuracy","quote":"The GBM achieved a significantly higher overall accuracy (0.89) compared to Logis-tic Regression (0.79), suggesting a more effective classification of the outcomes. For the prediction of reaching zero (Class 0), GBM exhibits higher precision (0.9142 vs. 0.8805) and notably improved recall (0.8886 vs. 0.7775), indicating a better ability 20 to correctly identify trajectories that eventually reach zero while","support":"full"},{"finding":"A metastable region separates the two attractors, acting as a transient phase space where trajectories exhibit heightened sensitivity to stochastic perturbations.","kind":"theoretical","quote":"Secondly, attractors are separated with a metastable region, functioning as a transient phase space, where trajectories exhibit heightened sensitivity to stochastic perturbations. Between the two attractor basins lies a metastable region, functioning as a transient phase space, where trajectories exhibit heightened sensitivity to stochastic perturbations.","support":"full"},{"finding":"Lyapunov exponents are consistently small but positive, suggesting intermittent or low‑dimensional chaotic behavior rather than full chaos.","kind":"theoretical","quote":"24 Values of Lyapunov exponent are consistently positive, which is a charac-teristic signature of chaotic systems, but very small which does not make a conclusive proof for the chaos. The system appears to exhibit an intermittent chaotic behavior or low-dimensional chaos, where the market exhibits complex-ity and irregularity, before eventually settling into a non-chaotic state.","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":22,"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":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":"2026-11-14T19:48:13Z"},"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/"}