Statistics//bootstrap

Estimating the uncertainty of a statistic by resampling the data with replacement many times and looking at the spread.

Estimating the uncertainty of a statistic by resampling the data with replacement many times and looking at the spread.

Gives a confidence interval without assuming a formula. If the interval on a model's edge over a baseline crosses zero, the edge is not established.

Resample the independent unit: for grouped data (sessions, patients), resample whole groups, not rows, or the interval is falsely narrow.

Pairs naturally with cross-validation to put error bars on a fold-averaged score.