causal
Skipping out-of-sample tests leaves no evidence
Running a regression on one half of a dataset and deploying it without testing on the other half leaves you with no evidence whether it will hold up, and discretionary managers frequently skip this step.
causalpending
Speaker
Leigh DrogenEvidence Quote
“If you don't run that out-of-sample and you're just like, "Wow, look, the regression works in the first five years."”
Created: 11/19/2026, 8:46:10 AM
Your notes
Loading notes...