Research team releases and open-sources LOB generative model market impact benchmark
Get up to speed
A research team led by George Nigmatulin released the LOB-Impact Benchmark for generative Limit Order Book (LOB) models and open-sourced it to evaluate the realism of market impact in generative LOB models. Test results from this benchmark show that the market impact index of generative LOB models ranges from 0.63 to 1.21, slightly higher than the 0.5 of the traditional square-root law. No impact decay during the relaxation phase has been observed in the models so far, and the research team speculates that this may be related to the still relatively short context window of the models.
AI-generated from reports · 2 hr agoUpdated
Coverage timeline
Follow the coverage to understand different sides of the event.
- arXiv q-finLOB-Impact Benchmark: Emergent Market Impact in Generative Limit Order Book Models
This paper, accepted by ICAIF '26, proposes the LOB-Impact Benchmark to evaluate the realism of market impact in generative limit order book models.
Trend of this event
Not enough consecutive observations yet; trend is not drawn.