The bootstrap is a resampling method to estimate the sampling distribution of a statistic.

The bootstrap is a technique to estimate the sampling distribution of a statistic, based on resampling from a dataset, estimating the parameters from the re-sampled data, and comparing those estimates to the (known) values for the dataset itself. There are many variants of bootstrapping used in specialized analyses.

For an extensive review of the bootstrap see:

  • Horowitz, J.L. (2001) "The Bootstrap", Handbook of Econometrics, in: J.J. Heckman & E.E. Leamer (eds.), Handbook of Econometrics, Edition 1, Vol. 5, Chapter 52, pp.3159-3228
  • Efron, B. and Tibshirani, R.J. (1994) "An Introduction to the Bootstrap", Chapman & Hall/CRC Monographs on Statistics & Applied Probability