LOB-Impact Benchmark: Emergent Market Impact in Generative Limit Order Book Models
Do AI-generated stock market simulations get trading costs right?
Researchers tested whether AI models trained to predict stock market order flows also learn realistic market impact—the price movement caused by large trades. They found that impact does emerge naturally in these models, but with slightly stronger effects than observed in real markets, and the models failed to capture how markets recover after large trades, likely because they don't remember enough context.
Trading firms use AI simulations to test strategies before risking real money. If a simulator underestimates how much your trade will move the market, you'll deploy strategies that look profitable in testing but lose money in reality. This benchmark provides the first rigorous way to catch that error before it costs traders millions.