A state-space generative model trained on synthetic limit order book data can partially imitate trading agent behavior, matching some action distributions while underestimating cancellations, with results limited by high variance.
Agent-based modeling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
q-fin.CP 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Prospects of Imitating Trading Agents in the Stock Market
A state-space generative model trained on synthetic limit order book data can partially imitate trading agent behavior, matching some action distributions while underestimating cancellations, with results limited by high variance.