Determine the number of bootstrap resamples needed for stable estimates.
Bootstrap intervals need enough resamples for the tails to be well populated. Two thousand is the practical minimum for a 95 per cent interval, and ten thousand is standard for publication. Too few resamples makes the interval endpoints unstable between runs, which is a Monte Carlo artefact rather than genuine statistical uncertainty.
Bootstrap Sample Size
B ≈ α(1−α) ÷ precision²; each tail should contain at least about 50 resamples
B ≈ α(1−α) ÷ precision²; each tail should contain at least about 50 resamples Bootstrap intervals need enough resamples for the tails to be well populated. Two thousand is the practical minimum for a 95 per cent interval, and ten thousand is standard for publication.
Too few resamples makes the interval endpoints unstable between runs, which is a Monte Carlo artefact rather than genuine statistical uncertainty.
This calculator takes 3 inputs: Confidence level, Desired Monte Carlo precision, Original sample size. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.