by Dmitry Lesnik, Tobias Schaefer
Continue on SelfMadeQuant
Suggested foundations for financial machine learning and credit and counterparty risk. Specialist prerequisites vary by paper.
Preserve time order and avoid future information in training.
Establish an interpretable baseline before a flexible model.
Separate default probability, loss severity and exposure.
Read conditional default and exposure quantities.
Account for dependence and overlapping observations.
Examine which information contributes to predictions.
Study dependence between defaults.
Choose a reproducible strategy, specify the trading rule, and test what remains after costs and estimation uncertainty.