by David Easley, Kara Karpman, Sumanta Basu
Continue on SelfMadeQuant
Suggested foundations for market microstructure and execution and financial machine learning. Specialist prerequisites vary by paper.
Read bid, ask, depth and queue-based market state.
Understand why fills and order flow can carry information.
Preserve time order and avoid future information in training.
Establish an interpretable baseline before a flexible model.
Connect trading speed and size to execution cost.
Measure execution relative to the decision price.
Account for dependence and overlapping observations.
Learn the trading state and execution trade-off before evaluating reinforcement-learning market makers.
Choose a reproducible strategy, specify the trading rule, and test what remains after costs and estimation uncertainty.