BOOTSTRAP (RESAMPLING)

The bootstrap method is a resampling technique used to estimate the distribution of a statistic by repeatedly sampling with replacement from the observed data. It allows for assessing the accuracy of statistical estimates and constructing confidence intervals without relying on parametric assumptions

Introduction to bootstrapping

What is bootstrap?

Explore the bootstrap method, a fundamental tool in statistics for resampling data and estimating the accuracy of sample estimates.

Jackknife method

Explore the Jackknife method, a resampling technique for bias correction and variance estimation that predates the bootstrap.

Bootstrap confidence intervals

Compare the four main bootstrap confidence interval methods and learn which one to use for accurate inference.

Parametric bootstrap

Parametric bootstrap

The parametric bootstrap fits a distributional model to the data and generates bootstrap samples from that fitted model rather than from the original observations.

Non-parametric bootstrap

Uniform bootstrap

Explore the uniform bootstrap method, a powerful tool in resampling that helps estimate the accuracy of sample statistics.

Smoothed bootstrap

Explore the smoothed bootstrap method, a resampling technique that adds kernel smoothing to better estimate continuous distributions from small samples.