normative

Sample Splitting Protects Against Over-Fitting

To protect against finding spurious patterns when searching many variables, use sample splitting: use one dataset to figure out the right model and a separate, clean dataset to estimate effects, so that you gain the better functional form from data mining while still obtaining correct standard errors.

normativepending

Speaker

Susan Athey

Evidence Quote

“you use one data set to figure out what the right model is; and another data set to actually estimate your effects.”

Source

Susan Athey on Machine Learning, Big Data, and Causation— EconTalk
Created: 6/13/2026, 12:14:45 AM

Your notes

Loading notes...