by Howard Caulfield, James P. Gleeson
Continue on SelfMadeQuant
Suggested foundations for financial time series and financial machine learning. Specialist prerequisites vary by paper.
Check whether a relationship can be estimated consistently over time.
Establish an interpretable baseline before a flexible model.
Preserve time order and avoid future information in training.
Study long-run relationships without confusing them with correlation.
Check the assumptions behind reported estimates.
Account for dependence and overlapping observations.
Choose a reproducible strategy, specify the trading rule, and test what remains after costs and estimation uncertainty.