by Anastasis Kratsios, Blanka Horvath, Florian Krach, Frank Rudzicz, Jacob-Junqi Tian, John Willes, Raeid Saqur, Yannick Limmer
Continue on SelfMadeQuant
Suggested foundations for financial machine learning and probability and statistical reasoning. Specialist prerequisites vary by paper.
Preserve time order and avoid future information in training.
Establish an interpretable baseline before a flexible model.
Read the distributions and expectations used in probabilistic arguments.
Interpret expectations after information is observed.
Account for dependence and overlapping observations.
Examine which information contributes to predictions.
Assess uncertainty rather than treating an estimate as a fact.
Choose a reproducible strategy, specify the trading rule, and test what remains after costs and estimation uncertainty.
Move from conditioning and expected values to timed problems with a defensible solution.