Bootstrap CI methods — percentile vs BCa vs studentized.

hard

Answer

$Percentile: use 2.5th and 97.5th quantiles of bootstrap distribution — simple but biased when the sampling distribution is skewed.$
$BCa (bias-corrected & accelerated): corrects bias and skewness \to generally best default.$
$Studentized (bootstrap-t): asymptotically most accurate but requires an SE estimate per bootstrap sample.$
$Q_{2.5}$
$Rule: default BCa; percentile only if you know the sampling distribution is symmetric.$

Check yourself — multiple choice

Only percentile
Percentile (biased under skew) / BCa (bias + skew corrected — default) / studentized (best asymptotically) / basic — BCa is safest default
Same as t-CI
Only Bayesian

Bootstrap CI: BCa default; percentile only if symmetric; studentized most accurate but costly.

#bootstrap#confidence-interval

Practise Statistics Fundamentals

215 interview questions in this topic.

Read the lesson Flashcards Take the quiz

Related questions