Conceptsample splitting
Sample Splitting
A statistical inference technique that divides data into a training subset and an inference subset, avoiding the overfitting bias that arises when the same batch of data is used both to estimate nuisance parameters and to construct test statistics.
Timeline (last 90 days)
Oct 15
Although sample splitting guarantees the theoretical properties, it reduces the sample size used to compute the test statistic, which may weaken test power in small-sample settings.
Unverified50%