How hard is it to pick the right model? MCS and backtest overfitting

DOI10.3233/AF-180231MaRDI QIDQ4586464zbMATH OpenOpenAlexWikidataFDO

Authors Diego Aparicio, Marcos López de Prado

Publication date 13 September 2018

Published in Algorithmic Finance (Search for Journal in Brave)

Full work available at URL https://doi.org/10.3233/af-180231

zbMATH Keywords

model selectionforecastingmachine learningmultiple testingmodel confidence set

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Financial applications of other theories (91G80)

Recommendations

• The Model Confidence Set
• Model selection confidence sets by likelihood ratio testing
• On over-fitting in model selection and subsequent selection bias in performance evaluation
• Comparison of value-at-risk models using the MCS approach
• On model selection from a finite family of possibly misspecified time series models

This page was built for publication: How hard is it to pick the right model? MCS and backtest overfitting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4586464)