Non-parametric bootstrapping

Non-parametric bootstrapping is a statistical method used to estimate the variability or uncertainty of a data-derived summary (like an average or median) without assuming a specific underlying distribution. It works by repeatedly resampling the original data with replacement—meaning some data points may be chosen multiple times—forming "new" sample sets. By calculating the summary statistic for each resampled set, we can gauge how much that statistic might vary if we repeated the process, providing insight into its reliability. Essentially, it uses the data itself to understand its own uncertainty, without relying on pre-defined models.