TRADES: Generating Realistic Market Simulations with Diffusion Models
TRADES is a transformer-based denoising diffusion model that produces realistic limit order book market simulations for financial research.
What it examines
The paper introduces TRADES, a transformer-based diffusion model that generates realistic limit order book simulations. It addresses the need for responsive market simulations for testing trading strategies and market impact experiments by modeling order flows using past market state. The work also provides an open-source framework and a synthetic dataset.
What it concludes
TRADES generates realistic market order flows close to real data, outperforming current methods. Its simulations can test trading strategies, market impact, and regulatory changes. The study suggests broad applications in finance research and strategy evaluation, while future work should refine evaluation protocols and validate simulation responsiveness.
Evidence objects
The paper introduces TRADES, a transformer-based denoising diffusion model that simulates market dynamics by replicating limit order book behavior, and offers DeepMarket, an open-source Python framework with a TRADES-LOB dataset.
key_findings bullet 1 · key_findings · validation V0
Employing self-attention on recent orders and LOB snapshots, TRADES accurately simulates market order flows and enhances prediction accuracy, closing simulation-real data gaps for Tesla and Intel by $3.27$ and $3.48$.
key_findings bullet 2 · key_findings · validation V0
Comprehensive experiments with principal component analysis, ablation studies, and market impact tests confirm TRADES' superiority over IABS and CGAN, although computational efficiency challenges persist and wider market validation remains essential.
key_findings bullet 3 · key_findings · validation V0
This paper innovates electronic financial market simulation using limit order book modeling, introducing a transformer-based denoising diffusion probabilistic engine (TRADES) as an original alternative to GANs. It effectively overcomes mode collapse and unstable training issues, presenting a novel quantitative evaluation metric and open-source framework that remarkably enhances market simulation research.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
Abstract: Financial markets are complex systems characterized by high statistical noise, nonlinearity, and constant evolution. Thus, modeling them is extremely hard. We address the task of generating realistic and responsive Limit Order Book (LOB) market simulations, which are fundamental for calibrating and testing trading strategies, performing market impact experiments, and generating synthetic market da… ▽ More Financial markets are complex systems characterized by high statistical noise, nonlinearity, and constant evolution. Thus, modeling them is extremely hard. We address the task of generating realistic and responsive Limit Order Book (LOB) market simulations, which are fundamental for calibrating and testing trading strategies, performing market impact experiments, and generating synthetic market data. Previous works lack realism, usefulness, and responsiveness of the generated simulations. To bridge this gap, we propose a novel TRAnsformer-based Denoising Diffusion Probabilistic Engine for LOB Simulations (TRADES). TRADES generates realistic order flows conditioned on the state of the market, leveraging a transformer-based architecture that captures the temporal and spatial characteristics of high-frequency market data. There is a notable absence of quantitative metrics for evaluating generative market simulation models in the literature. To tackle this problem, we adapt the predictive score, a metric measured as an MAE, by training a stock price predictive model on synthetic data and testing it on real data. We compare TRADES with previous works on two stocks, reporting an x3.27 and x3.47 improvement over SoTA according to the predictive score, demonstrating that we generate useful synthetic market data for financial downstream tasks. We assess TRADES's market simulation realism and responsiveness, showing that it effectively learns the conditional data distribution and successfully reacts to an experimental agent, giving sprout to possible calibrations and evaluations of trading strategies and market impact experiments. We developed DeepMarket, the first open-source Python framework for market simulation with deep learning. Our repository includes a synthetic LOB dataset composed of TRADES's generates simulations. We release the code at github.com/LeonardoBerti00/DeepMarket. △ Less
Source row: 2061 · abstract type: unknown