Can an existing bootstrap distribution support a new confidence level without rerunning the expensive statistic?
StatisticsLevel: IntermediateBootstrap uncertainty
Can an existing bootstrap distribution support a new confidence level without rerunning the expensive statistic?
Hints
- Quantile choice and simulation are separate steps.
- Do not reuse results from a different dataset.
Answer
Yes, reuse the stored distribution through bootstrap_result and request zero additional resamples. Verify the original data and statistic are unchanged.
Source and licence: SciPy — resampling and Monte Carlo inference · BSD-3-Clause · answer: Quantoros editorial