Paper: arXiv 2311.12247
Authors: Nicolas Cofre, Magdalena Mosionek-Schweda
Abstract
This paper presents an agent based model of an electronic market with two types of trading agents. One type follows a mean reverting strategy and the other, the speculative trader, tracks the maximum realised return over recent trades. The speculators have a distribution of returns concentrated on negative returns, with a small fraction making profits. The market experiences an increased volatility and prices that greatly depart from the fundamental value of the asset. Our research provides synthetic datasets of the order book to study its dynamics under different levels of speculation
Complexity vs Empirical Score
- Math Complexity: 4.5/10
- Empirical Rigor: 6.0/10
- Quadrant: Street Traders — empirical and applied, lower math complexity
Why this score: The paper relies on established agent-based modeling and stochastic processes (e.g., OU, Poisson, exponential distributions) rather than novel advanced mathematics, placing it in the lower-mid range for math complexity. It is highly implementation-heavy, providing specific code algorithms and describing synthetic order book datasets for replication, which indicates strong empirical rigor for a simulation study.
Research Flowchart
flowchart TD A["Research Goal"] --> B["Define Agents & Strategies"] B --> C["Set Market Rules & Parameters"] C --> D["Run Simulations"] D --> E["Generate Synthetic Order Book Data"] E --> F["Analyze Price, Vol & Bubble Formation"] F --> G["Key Outcomes"] style A fill:#f9f,stroke:#333,stroke-width:2px style G fill:#9f9,stroke:#333,stroke-width:2px
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes an agent‑based model of an electronic market with mean‑reverting and speculative trading agents and provides synthetic order‑book datasets to study market dynamics under varying levels of speculation.
Source sentence
This paper presents an agent based model of an electronic market with two types of trading agents. Our research provides synthetic datasets of the order book to study its dynamics under different levels of speculation
Assumes 6
- The asset fundamental value follows an Ornstein‑Uhlenbeck mean‑reverting process. model
Source sentence
2.3. The asset fundamental value The asset fundamental value follows an OU mean-reverting process which is affected by additional shocks.
- Fundamental value shocks are drawn from a mixture of two equally weighted normal distributions with opposite means. model
Source sentence
Those shocks follow a mixture of 2 equally weighted normal distributions, with the same absolute mean, representing posi-tive and negative shocks to the value of the asset.
- Agents receive noisy observations of the fundamental value, normally distributed around the true value. data
Source sentence
The agents can get noisy observations of the value that are normally distributed around the true fundamental value, which creates heterogeneity about the value beliefs.
- Speculative agents base their trading decisions on observed returns rather than on fundamental‑value beliefs. model
Source sentence
The speculative agents base their trading strategies on the observed returns and not on the fundamental value beliefs. 3. Results 3.1.
Show 2 more
- Inter‑trade times for each agent are exponentially distributed. execution
Source sentence
There is heterogeneity regarding the timing of the trading, as each agent has a time between trades exponentially distributed.
- The window length for maximum‑return calculation of speculators is Poisson distributed. model
Source sentence
This number of periods is Poisson distributed.
Data
- Synthetic order‑book dataset synthetic unknown
Source sentence
Our research provides synthetic datasets of the order book to study its dynamics under different levels of speculation
Costs
Not applicable
Finds 5
- Speculators have a return distribution concentrated on negative returns, with only a small fraction making profits. statistical
Source sentence
The speculators have a distribution of returns concentrated on negative returns, with a small fraction making profits.
- The market exhibits increased volatility and prices that deviate strongly from the fundamental value. performance
Source sentence
The market experiences an increased volatility and prices that greatly depart from the fundamental value of the asset.
- During the first half of the simulation, speculative agents trade little while mean‑reverting agents generate increased volatility. performance
Source sentence
there is almost no trading done by the speculators and there is an increased volatility among the trades executed by the 100 mean reverting agents.
- In the second half, speculative agents become active, causing another period of high volatility that amplifies demand and market‑maker inventory imbalance. performance
Source sentence
In the second half of the simulation, the speculative agents start to engage in trading, producing another period of high volatility. This increased volatility promotes the entrance of more speculative agents, which in turns increases the demand for the asset, creating an imbalance in the inventory of the market maker.
- Speculative activity builds up several timesteps before the bubble peak, suggesting a window for early detection. theoretical
Source sentence
In our simulated market, the speculative activity builds up several timesteps before the peak of the bubble.
Checked by others
Supported by
- News-driven Expectations and Volatility Clustering
Compare the findings
This paper: In the second half, speculative agents become active, causing another period of high volatility that amplifies demand and market‑maker inventory imbalance.
That paper: Excess volatility arises from the amplifying feedback of speculative trading.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (43)
- Lottery demand and the asset growth anomaly (2022) — link
- Speculation and lottery-like demand in cryptocurrency markets (2021) — link
- Critical comparison of several order-book models for stock-market fluctuations (2008) — link
- Individual speculative behavior and overpricing in experimental asset markets (2019) — link
- ABIDES: Towards High-Fidelity Market Simulation for AI Research (2019) — link
- ABIDES-Gym: Gym Environments for Multi-Agent Discrete Event Simulation and Application to Financial Markets (2021) — link
- A Synthetic Limit Order Book Dataset for Benchmarking Forecasting Algorithms under Distributional Shift (2022)
- Time-varying demand for lottery: Speculation ahead of earnings announcements (2020) — link
and 35 more
- Scenario-Based Synthetic Dataset Generation for Mobile Money Transactions (2022) — link
- Get Real: Realism Metrics for Robust Limit Order Book Market Simulations (2021) — link
- A mathematical approach to order book modeling (2013)
- How to Evaluate Trading Strategies: Single Agent Market Replay or Multiple Agent Interactive Simulation? (2019) — link
- High frequency trading strategies, market fragility and price spikes: an agent based model perspective (2019)
- Market making and mean reversion (2011)
- Mathematical modeling for market making and related problems of financial liquidity: a song of assets and traders. (2021)
- Market makers activity: behavioural and agent based approach (2022)
- Market making by an FX dealer: tiers, pricing ladders and hedging rates for optimal risk control (2021) — link
- The digital pound: a new form of money for households and businesses? (2023)
- Algorithms and Collusion: Competition Policy in the Digital Age (2017)
- The bubble game: An experimental study of speculation (2013)
- Gymnasium (2023) — link
- Asset bubbles and overlapping generations (1985)
- Speculative overpricing in asset markets with information flows (2012)
- Maxing out: Stocks as lotteries and the cross-section of expected returns (2011)
- Testing for explosive bubbles in the presence of non-Gaussian conditions (2023)
- Bubbles and crashes (2003)
- Bubbles and financial professionals (2020)
- Testing for explosive bubbles in the presence of autocorrelated innovations (2020)
- Rational bubbles: Too many to be true? (2023)
- Using covariates to improve the efficacy of univariate bubble detection methods (2023)
- Autonomous algorithmic collusion: Q-learning under sequential pricing (2021)
- Testing for short explosive bubbles: A case of Brent oil futures price (2023)
- Getting at systemic risk via an agent-based model of the housing market (2012)
- Agent-based modeling in economics and finance: Past, present, and future (2022)
- Agent-based computational finance: Suggested readings and early research (2000)
- Advances in the agent-based modeling of economic and social behavior (2021)
- Agent-based modeling and investors’ behavior explanation of asset price dynamics on artificial financial markets (2014)
- Agent-based models in financial market studies (2018)
- Behavioral agent-based framework for interacting financial markets (2020)
- Development of an agent-based speculation game for higher reproducibility of financial stylized facts (2019)
- An agent-based model of financial market efficiency dynamics (2022)
- Agent-based models of financial markets: A comparison with experimental markets (2001)
- Reprint of: Do retail traders destabilize financial markets? An investigation surrounding the COVID-19 pandemic (2023)
0 of 43 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.