Paper: arXiv 2508.08698
Authors: Zhuohan Wang, Carmine Ventre
Abstract
Modeling limit order books (LOBs) dynamics is a fundamental problem in market microstructure research. In particular, generating high-dimensional volume snapshots with strong temporal and liquidity-dependent patterns remains a challenging task, despite recent work exploring the application of Generative Adversarial Networks to LOBs. In this work, we propose a conditional \textbf{“Diff”}usion model for the generation of future LOB \textbf{“Volume”} snapshots (\textbf{“DiffVolume”}). We evaluate our model across three axes: (1) \textit{“Realism”}, where we show that DiffVolume, conditioned on past volume history and time of day, better reproduces statistical properties such as marginal distribution, spatial correlation, and autocorrelation decay; (2) \textit{“Counterfactual generation”}, allowing for controllable generation under hypothetical liquidity scenarios by additionally conditioning on a target future liquidity profile; and (3) \textit{“Downstream prediction”}, where we show that the synthetic counterfactual data from our model improves the performance of future liquidity forecasting models. Together, these results suggest that DiffVolume provides a powerful and flexible framework for realistic and controllable LOB volume generation.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 7.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced diffusion model theory with SDEs and detailed mathematical derivations, indicating high math complexity; it also presents concrete evaluation on realism, counterfactual generation, and downstream prediction tasks, suggesting strong empirical rigor.
Research Flowchart
flowchart TD
A["Research Goal: Generate<br>high-fidelity LOB volume snapshots<br>with temporal/liquidity patterns"] --> B["Data: Historical LOB volume data<br>+ Time of Day + Liquidity Profiles"]
B --> C["Methodology: Conditional Diffusion Model<br>DiffVolume"]
C --> D["Generative Process:<br>Samples future volume snapshots<br>conditioned on past history"]
D --> E{"Key Outcomes"}
E --> F["1. Realism: Superior statistical<br>matches to marginal/spatial/autocorrelation"]
E --> G["2. Counterfactuals: Controllable generation<br>under hypothetical liquidity scenarios"]
E --> H["3. Downstream: Synthetic data improves<br>future liquidity forecasting models"]
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 DiffVolume, a conditional diffusion model for generating future limit order book volume snapshots.
Source sentence
In this work, we propose a conditional Diffusion model for the generation of future LOB Volume snapshots (DiffVolume).
Assumes 3
- The model conditions on past volume trajectories and time of day to capture spatial and temporal dependencies. model
Source sentence
We develop a novel diffusion model architecture conditioning on past volume trajectories and time of day, which captures the spatial correlation structures and intricate temporal dependencies presented in LOB volume data
- Stocks are classified by tick size (big, medium, small) which influences order‑book dynamics. market
Source sentence
Stocks are classified as big, medium, or small tick size according to how their tick size compares to the typical bid-ask spread.
- The first and last 30 minutes of each trading session are excluded to avoid non‑stationary behavior. data
Source sentence
To mitigate the non-stationary behavior typically observed in limit order books—particularly around market open and close—we exclude the first and last 30 minutes of each continuous trading sessio…
Data
Costs
Not applicable
Finds 3
- DiffVolume better reproduces statistical properties such as marginal distribution, spatial correlation, and autocorrelation decay compared to baselines. statistical
Source sentence
DiffVolume, conditioned on past volume history and time of day, better reproduces statistical properties such as marginal distribution, spatial correlation, and autocorrelation decay
- The model enables controllable counterfactual generation under hypothetical liquidity scenarios by conditioning on a target future liquidity profile. other
Source sentence
Counterfactual generation, allowing for controllable generation under hypothetical liquidity scenarios by additionally conditioning on a target future liquidity profile
- Augmenting training data with synthetic samples from DiffVolume consistently improves short‑term liquidity forecasting performance. predictive-accuracy
Source sentence
By augmenting the training data with generated samples, we achieve consistent improvements in forecasting short-term liquidity.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 21 related papers). Coverage is partial, so this is neither replication nor refutation.
2 papers in the archive
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 (85)
- Score-based generative modeling through stochastic differential equations (2021)
- Generative modeling by estimating gradients of the data distribution (2019)
- Denoising diffusion probabilistic models (2020)
- CSDI: Conditional score-based diffusion models for probabilistic time series imputation (2021)
- Diffusion models for adversarial purification (2022) — link
- Can Machines’ Learn’Finance? (2020)
- Deep unsupervised learning using nonequilibrium thermodynamics (2015)
- Improved denoising diffusion probabilistic models (2021)
and 77 more
- Denoising diffusion implicit models (2021)
- A connection between score matching and denoising autoencoders (2011)
- Elucidating the design space of diffusion-based generative models (2022)
- Diffusion models for time-series applications: a survey (2023)
- Autoregressive denoising diffusion models for multivariate probabilistic time series forecasting (2021)
- Scoregrad: Multivariate probabilistic time series forecasting with continuous energy-based generative models (2021) — link
- Generative time series forecasting with diffusion, denoise, and disentanglement (2022)
- Non-autoregressive conditional diffusion models for time series prediction (2023)
- Diffusion-based time series imputation and forecasting with structured state space models (2022) — link
- Regular time-series generation using sgm (2023) — link
- On the constrained time-series generation problem (2024)
- Forecasting stock market indices using padding-based fourier transform denoising and time series deep learning models (2021)
- Stock price prediction via discovering multi-frequency trading patterns (2017)
- Denoised Labels for Financial Time Series Data via Self-Supervised Learning (2022)
- Extracting and composing robust features with denoising autoencoders (2008)
- Prediction of the change points in stock markets using DAE-LSTM (2021)
- Deep Learning-based Integrated Framework for stock price movement prediction (2022)
- The Fourier transform and its applications (1986)
- Architectures for wavelet transforms: A survey (1996)
- The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis (1998)
- A hybrid model for financial time series forecasting—integration of EWT, ARIMA with the improved ABC optimized ELM (2020)
- Forecasting East Asian indices futures via a novel hybrid of wavelet-PCA denoising and artificial neural network models (2016)
- A novel time-series model based on empirical mode decomposition for forecasting TAIEX (2014)
- Forecasting wavelet neural hybrid network with financial ensemble empirical mode decomposition and MCID evaluation (2021)
- Understanding diffusion models: A unified perspective (2022) — link
- Reverse-time diffusion equation models (1982)
- Modeling financial time-series with generative adversarial networks (2019)
- Empirical properties of asset returns: stylized facts and statistical issues (2001)
- Enriching financial datasets with generative adversarial networks (2019)
- Wasserstein generative adversarial networks (2017)
- Time-series generative adversarial networks (2019)
- Synthetic Data Augmentation for Deep Reinforcement Learning in Financial Trading (2022)
- Wavenet: A generative model for raw audio (2016) — link
- Conditional sig-wasserstein gans for time series generation (2020) — link
- Using generative adversarial networks to synthesize artificial financial datasets (2020) — link
- On convergence and stability of gans (2017) — link
- Generating realistic stock market order streams (2020)
- Total variation regularization for image denoising, I. Geometric theory (2008)
- Wavelet and FFT Based Image Denoising Using Non-Linear Filters. (2015)
- From the bird’s eye to the microscope: A survey of new stylized facts of the intra-daily foreign exchange markets (1997)
- Lightgbm: A highly efficient gradient boosting decision tree (2017)
- Learning multiple stock trading patterns with temporal routing adaptor and optimal transport (2021)
- Listening to chaotic whispers: A deep learning framework for news-oriented stock trend prediction (2018)
- Stock movement prediction from tweets and historical prices (2018)
- Quant GANs: deep generation of financial time series (2020)
- Corrgan: Sampling realistic financial correlation matrices using generative adversarial networks (2020)
- Abides: Towards high-fidelity market simulation for AI research (2019) — link
- Efficiently modeling long sequences with structured state spaces (2021) — link
- Attention is all you need (2017)
- Guided diffusion model for adversarial purification (2022) — link
- DiffsFormer: A Diffusion Transformer on Stock Factor Augmentation — → 2402.06656
- Classifier-free diffusion guidance (2021)
- Diffusion models beat gans on image synthesis (2021)
- Styletime: Style transfer for synthetic time series generation (2022)
- Trades, quotes and prices: financial markets under the microscope (2018)
- Limit order books (2013)
- Towards realistic market simulations: a generative adversarial networks approach (2021)
- Generative AI for end-to-end limit order book modelling: A token-level autoregressive generative model of message flow using a deep state space network — → 2309.00638
- Limit order book simulation with generative adversarial networks (2023)
- A stochastic model for order book dynamics (2010)
- Hawkes processes in finance (2015)
- Modelling financial high frequency data using point processes (2009)
- An agent based model of the E-Mini S&P 500 applied to Flash Crash analysis (2012)
- Quantifying reflexivity in financial markets: Toward a prediction of flash crashes (2012)
- ABIDES-gym: gym environments for multi-agent discrete event simulation and application to financial markets (2021)
- How to evaluate trading strategies: Single agent market replay or multiple agent interactive simulation? (2019) — link
- A generative model of a limit order book using recurrent neural networks (2023)
- Film: Visual reasoning with a general conditioning layer (2018)
- Lobster: Limit order book reconstruction system (2011)
- HLOB–Information persistence and structure in limit order books — → 2405.18938
- Deep learning (2016)
- Adam: A Method for Stochastic Optimization (2015)
- A Financial Time Series Denoiser Based on Diffusion Models (2024)
- Diffusion variational autoencoder for tackling stochasticity in multi-step regression stock price prediction — → 2309.00073
- Denoising Diffusion Probabilistic Model for Realistic Financial Correlation Matrices (2024)
- WaveNet: A Generative Model for Raw Audio (2016)
- Tutorial on diffusion models for imaging and vision (2024)
4 of 85 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.