Bootstrapping refers to the general procedure of sampling with replacement from a sample of data. The original sample of data is considered the population, and a new sample of data is generated by randomly sampling with replacement from the original sample. This new sample is called a bootstrap sample. Based on this bootstrap sample, a statistic is calculated and recorded. Then the process is repeated. Another bootstrap sample is obtained, and the statistic is calculated again and recorded. Typically, this process is repeated many times such that the number of bootstrap samples (B) is large (e.g., B = 1,000). Overall, bootstrapping provides a general approach to understanding the characteristics of a population by using the statistics that were calculated on each of the bootstrap samples.

The ...

  • 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