normative

Sample splitting protects against overfitting

To get valid statistical inference after data-driven model selection, use sample splitting: use one half of the data to choose the model and the other half to estimate the effect; though throwing away data seems horrifying in a small-data world, the gain in functional-form fit justifies it and yields a correct standard error.

normativepending

Speaker

Susan Athey

Evidence Quote

“you need to find a clean data set to actually get a correct standard error.”

Source

Susan Athey on Machine Learning, Big Data, and Causation— EconTalk
Created: 6/13/2026, 3:35:37 AM

Your notes

Loading notes...