It is well-know that for heavy-tailed distributions the bootstrap can lead to inconsistent estimation of the distribution of the sample mean; and that this difficulty may be overcome by using the ...

Several uncertainty metrics can be computed solely based on the bootstrap distribution of these low-dimensional coordinates, without calculating or storing the p-dimensional bootstrap components.

The last part, the non-parametric bootstrap, is simply a way to estimate the probability distribution of events via sampling over and over a subset of the actual empirical data.