Permutation Tests and Resampling Techniques

  • David S. Salsburg
Part of the Springer Series in Statistics book series (SBH)


Recall that in significance testing we observe data and propose as a null hypothesis a statistical model that could have generated such data, lacking a treatment effect. We then calculate the probability of the observed data. If that probability is very low, we conclude that the null hypothesis is in adequate to describe how the data were generated. To calculate the probability, we consider a set of possible observations:

The data we observed plus all possible patterns of data that would have been “more extreme” than these data.


Permutation Test Percent Coverage Randomization Pattern Percent Confidence Interval Empirical Distribution Function 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag New York, Inc. 1992

Authors and Affiliations

  • David S. Salsburg
    • 1
  1. 1.Pfizer Research DivisionPfizer, Inc.GrotonUSA

Personalised recommendations