Paper: arXiv 2312.16190

Authors: Raffaele Giuseppe Cestari, Filippo Barchi, Riccardo Busetto, Daniele Marazzina, Simone Formentin

Abstract

Accurately forecasting the direction of financial returns poses a formidable challenge, given the inherent unpredictability of financial time series. The task becomes even more arduous when applied to cryptocurrency returns, given the chaotic and intricately complex nature of crypto markets. In this study, we present a novel prediction algorithm using limit order book (LOB) data rooted in the Hawkes model, a category of point processes. Coupled with a continuous output error (COE) model, our approach offers a precise forecast of return signs by leveraging predictions of future financial interactions. Capitalizing on the non-uniformly sampled structure of the original time series, our strategy surpasses benchmark models in both prediction accuracy and cumulative profit when implemented in a trading environment. The efficacy of our approach is validated through Monte Carlo simulations across 50 scenarios. The research draws on LOB measurements from a centralized cryptocurrency exchange where the stablecoin Tether is exchanged against the U.S. dollar.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced stochastic processes (Hawkes models) with detailed mathematical formulations and derivations, indicating high mathematical complexity. It is heavily data-driven with specific empirical validation using real limit order book data, Monte Carlo simulations, and reported trading performance, showing high empirical rigor.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Predict cryptocurrency return signs using LOB data"] --> B{"Methodology"}
  B --> C["Hawkes Process Model"]
  B --> D["Continuous Output Error COE Model"]
  C & D --> E["Data Input<br>Cryptocurrency LOB Data Tether USD"]
  E --> F["Computational Process<br>Monte Carlo Simulations 50 Scenarios"]
  F --> G["Key Findings Outcomes"]
  G --> H["Superior Prediction Accuracy vs Benchmarks"]
  G --> I["Higher Cumulative Profit in Trading"]

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 a Hawkes‑based return sign prediction model that estimates the next LOB order time and combines it with a Continuous Output Error (COE) model to forecast return signs.

Source sentence
For the first time, as far as we are aware, we develop a return sign prediction model based on Hawkes processes to estimate the next LOB order time and we couple the predicted information with a COE model to predict the return sign.

Assumes 4

  • Existence of a Continuous Output Error (COE) model linking base imbalance to future returns. model
    Source sentence
    We assume the existence of a Continuous Output Error (COE) model that connects BI and future returns as follows:
  • Zero‑return events are removed from the sample. data
    Source sentence
    In particular, we stick to remove zero return events, possibly losing information content.
  • No transaction costs are incurred in the trading simulation. execution
    Source sentence
    No transaction costs are considered in this exercise.
  • The Oracle benchmark assumes perfect knowledge of the next event time. model
    Source sentence
    Oracle algorithm, when combined with the COE model, replicates the process performed in Busetto and Formentin (2023), but with our dataset. In essence, it assumes knowledge of the timing of the next event.

Data

  • Tether (USDT) LOB dataset order-book unknown
    USDT/USD on a centralized exchange
    Source sentence
    The research draws on LOB measurements from a centralized cryptocurrency exchange where the stablecoin Tether is exchanged against the U.S. dollar.
  • Real cryptocurrency LOB data order-book unknown
    Source sentence
    We apply this prediction strategies on real cryptocurrency data collected in a LOB dataset.

Costs

Not modeled A trading or execution result is reported without costs.

Source sentence
No transaction costs are considered in this exercise.

Finds 3

  • The Hawkes‑based strategy outperforms benchmark algorithms in return‑sign accuracy, approaching the performance of an ideal Oracle. performance
    Source sentence
    Hawkes outperforms the Naive and MA due to its machine learning-based prediction method for the next event time.
  • In trading simulations, the Hawkes strategy yields higher total profit than Naive and Moving Average benchmarks, closely matching the Oracle’s profit. performance
    Source sentence
    Hawkes shows satisfactory performances, approaching ideal performances—next, the MA and Naive benchmarking strategies.
  • Base imbalance (BI) exhibits a strong negative correlation with returns (ρ = −0.96), confirming its usefulness as a regressor for cryptocurrency LOB data. statistical
    Source sentence
    ρ = −0.96. This result, shows the promising relation between BI and R, confirming the intuition of Busetto and Formentin (2023) of the relevance of base imbalance feature as meaningful regressor for return sign prediction.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 35 related papers). Coverage is partial, so this is neither replication nor refutation.

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 (31)

  1. Continuous-time modeling of financial returns based on Limit Order Book data (2023)
  2. Buy low, sell high: A high frequency trading perspective (2014)
  3. The information content of an open limit-order book (2009)
  4. Predictive Methods for H awkes Processes in High Frequency Finance (2022)
  5. State-dependent H awkes processes and their application to limit order book modelling (2022)
  6. Modelling Limit Order Book Dynamics Using H awkes Processes (2017)
  7. High-dimensional H awkes processes for limit order books: modelling, empirical analysis and numerical calibration (2018)
  8. Spectra of some self-exciting and mutually exciting point processes (1971)
and 23 more
  1. A cluster process representation of a self-exciting process (1974)
  2. The information content of the limit order book: evidence from NYSE specialist trading decisions (2005)
  3. Price jump prediction in limit order book (2012) — link
  4. Forecasting stock prices from the limit order book using convolutional neural networks (2017)
  5. The role of volume in order book dynamics: a multivariate H awkes process analysis (2017)
  6. Modeling foreign exchange market activity around macroeconomic news: H awkes-process approach (2015)
  7. Hawkes model for price and trades high-frequency dynamics (2014)
  8. A time-modulated H awkes process to model the spread of COVID-19 and the impact of countermeasures (2021)
  9. Inference of functional connectivity in neurosciences via H awkes processes (2013)
  10. Flexible spatio-temporal H awkes process models for earthquake occurrences (2023)
  11. Longitudinal modeling of social media with H awkes process based on users and networks (2017)
  12. Stablecoins and cryptocurrency returns: What is the role of T ether? (2020)
  13. Cryptocurrency, a successful application of blockchain technology (2020)
  14. Forecasting cryptocurrency volatility (2022)
  15. Refined instrumental variable identification of continuous-time OE and BJ models from irregularly sampled data (2013)
  16. High Frequency Trading Strategies (2021)
  17. Hawkes processes (2015) — link
  18. An introduction to the theory of point processes: volume I: elementary theory and methods (2003)
  19. Tutorial on maximum likelihood estimation (2003)
  20. Deeplob: Deep convolutional neural networks for limit order books (2019)
  21. Hawkes process for understanding the influence of pathogenic social media accounts (2019)
  22. Cryptocurrencies and stablecoins: a high-frequency analysis (2022)
  23. Market impact and efficiency in cryptoassets markets (2023)

0 of 31 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.