Function Reference

The statistics-resampling package is an Octave package and Matlab toolbox that can be used to perform a wide variety of statistics tasks using non-parametric resampling methods. In particular, the functions included can be used to estimate bias, uncertainty (standard errors and confidence intervals), prediction error, information criteria, and calculate p-values for null hypothesis significance tests. Variations of the resampling methods are included that improve the accuracy of the statistics for small samples and samples with complex dependence structures.

Select category:

Cloned Matlab functions

bootstrp
Bootstrap: Resample with replacement to generate new samples and return the statistic(s) calculated by evaluating the specified function on each resample.
bootci
Advanced bootstrap confidence intervals: Performs bootstrap and calculates various types of confidence intervals from the resulting bootstrap statistics.

Main functions

bootknife
Performs one or two levels of bootknife resampling and calculates bootstrap bias, standard errors and confidence intervals.
bootclust
Performs balanced bootstrap (or bootknife) resampling of clusters or blocks of data and calculates bootstrap bias, standard errors and confidence intervals.
bootwild
Performs wild bootstrap and calculates bootstrap-t confidence intervals and p-values for the mean, or the regression coefficients from a linear model.
bootbayes
Performs Bayesian nonparametric bootstrap and calculates posterior statistics for the mean, or regression coefficients from a linear model.
randtest
Performs permutation or randomization tests for regression coefficients.
randtest1
Performs a permutation or randomization test to assess if a sample comes from a population with a value given for the mean or some other location parameter
randtest2
Performs a permutation or randomization test to compare the distributions of two independent or paired data samples.
bootlm
Uses bootstrap to calculate confidence intervals (and p-values) for the regression coefficients from a linear model and performs N-way ANOVA.
boot1way
Performs resampling under the null hypothesis and computes p-values for (multiple) comparisons among independent samples in a one-way layout.
bootmode
Uses bootstrap to evaluate the likely number of real peaks (i.e. modes) in the distribution of a single set of data.

Utility Functions

boot
This function returns resampled data or indices created by balanced bootstrap or bootknife resampling.
bootcdf
Computes the empirical cumulative distribution function (ECDF), accounting for the presence of ties.
bootint
Computes percentile confidence interval(s) directly from a vector (or row- major matrix) of bootstrap statistics.
credint
Computes credible interval(s) directly from a vector (or row-major matrix) of the posterior(s) obtained by bayesian bootstrap.
deffcalc
Computes the design effect (DEFF), which can subsequently be used to correct sample size calculations using the 'sampszcalc' function.
sampszcalc
Performs sample size calculations, with optional correction for the design effect deviating from unity.

Parameter functions

smoothmedian
Calculates a smoothed version of the median.
smoothmad
Calculates a smoothed version of the median absolute deviation (MAD).
cor
Vectorized function for computing Pearson's correlation coefficient (RHO) between each of the respective columns in two data vectors or matrices.

Package: statistics-resampling