Sample splitting separates data exploration from inference in target trial emulations, delivering valid coverage even after protocol selection informed by part of the data.
Let’s Take the Con Out of Econometrics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A simple strategy for valid inference in target trial emulations
Sample splitting separates data exploration from inference in target trial emulations, delivering valid coverage even after protocol selection informed by part of the data.