FlowLOB generative model uses flow matching for realistic order book simulation

Researchers have developed FlowLOB, a new generative model for limit order book (LOB) trajectories that utilizes flow matching. This model demonstrates improved realism and controllability compared to existing agent-based and deep generative simulators, particularly at finer sampling frequencies. FlowLOB also shows the ability to generalize to unseen financial instruments without retraining, outperforming diffusion models in terms of sampling efficiency. AI

IMPACT This research could lead to more accurate and efficient financial market simulations, aiding in strategy development and risk assessment.

RANK_REASON The cluster describes a new generative model presented in an arXiv paper, detailing its methodology and performance. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• FlowLOB
• Gotit.pub
• Hong Kong Exchanges and Clearing Limited
• Hugging Face
• Namid Stillman Dr.
• ScienceCast

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →