• Entry
  • Reader's guide
  • Entries A-Z
  • Subject index

Bootstrapping is an approach to properties of statistics, such as sampling variances, standard errors, and confidence intervals, that does not rely on a particular assumption about the shape of the distribution around a given statistic. Bootstrapping is therefore said to be a nonparametric approach to statistical inference. It can be particularly useful when the researcher does not know the theoretical distribution of a given test statistic or when no such distribution exists.

Bootstrap methods for evaluating statistics rely on data-based simulations wherein the observed data stand in for the population of interest. Measures of uncertainty around a statistic that are obtained via the bootstrap therefore might be thought of as being drawn from samples of a given sample, as bootstrapping is a computationally intensive method that ...

    • Loading...
    locked icon

    Sign in to access this content

    Get a 30 day FREE TRIAL

    • Watch videos from a variety of sources bringing classroom topics to life
    • Read modern, diverse business cases
    • Explore hundreds of books and reference titles