iboot package

(iboot is no longer being developed; consider using the newer Octave statistics-resampling package for Octave/MATLAB)

Package maintainer

Andrew Penn ([email protected])

Package contributors

Andrew Penn

Citations

If you use this package please include the following citation(s):

• Penn, Andrew Charles. (2020). iboot: Iterated bootstrap for small samples or samples with complex dependence structures. Zenodo. https://doi.org/10.5281/zenodo.3992392

A statistics package for Octave/Matlab providing a variety of bootstrap resampling tools

This package of functions can be used to estimate uncertainty (confidence intervals) and test hypotheses (p-values) using bootstrap. Variations of the bootstrap are included that improve the accuracy of bootstrap statistics for small samples and samples with complex dependence structures. The Octave statistics-bootstrap package is forked from the GitHub repository iboot.

Requirements and dependencies

This package is known to be compatible with versions of Octave v3.6.0+ and Matlab v7.4.0+.

Most features of of the package do not depend on other packages. An exception is the parallel computing options, which require either the Octave forge parallel package or the Parallel Computing Matlab Toolbox.

Installation

To install (or test) the iboot package at it's existing location in either Octave or Matlab, follow these steps:

• Download the package. If it is a compressed file, decompress it.
• Open Octave or Matlab command prompt.
• cd to the package directory. (The directory contains a file called 'install.m').
• Type install. The package will load now (and automatically in the future) when you start Octave/Matlab.

To uninstall, cd to the package directory and type uninstall.

Usage

Functions

• ibootci calculates confidence intervals (calibrated) by iterated bootstrap resampling.
• ibootp calculates a two-tailed p-value for hypothesized value of the statistic using bootstrap.
• iboottest is a convenience function that uses ibootci and ibootp to compute confidence intervals and p-values for the difference between two paired samples or between one sample and a population value (two-tailed). This function resamples under the alternative hypothesis.
• iboottest2 is a convenience function that uses ibootci and ibootp to compute confidence intervals and p-values for the difference between two independent (i.e. unpaired) samples. This function resamples under the alternative hypothesis.
• plotboot plots an overlay of a histogram, kernel density estimate and interval limits from bootstrap statistics.

At the Octave command prompt, type help function-name for more information about the function and it's usage.

Notes

Be aware that that many of the bootstrap functions in this package are deterministic through the setting of a random seed on each function call. This will be a problem if testing the bootstrap functions in simulations that depend on generating samples using Matlab's or Octave's random number generator. The solution is to set UseParallel in the paropt structure input argument to True, which disables the resetting of the random number generator.

Development roadmap

• Create more documentation and guidance for using the functions in this package