statistics-bootstrap-v3.7.7

bootknife no longer depends on functions within the statistics package/toolbox

statistics-bootstrap-v3.7.6

bootknife function - added expanded BCa percentile when bootfun is the mean

statistics-bootstrap-v3.7.5

Fixed bug in for double bootstrap in bootknife in Octave, where it attempts to call interp1q, which is a Matlab but not Octave function. This fix enhances performance in matlab too.

statistics-bootstrap-v3.7.4

bootknife bug fix

statistics-bootstrap-v3.7.3

New features to bootknife.m

• Added ability for bootknife to use bootfun that returns either scalar or a vector
• Added ability to calculate BCa intervals for stratified samples

statistics-bootstrap-v3.7.2

ibootnhst: Added ability to specify additional input arguments to bootfun

statistics-bootstrap-v3.7.1

Performance enhancement for ibootnhst. When the function of the (univariate) data is the mean (which is default), ibootnhst automatically avoids resampling in the calculation of the standard error of the mean.

statistics-bootstrap-v3.7.0

• bootknife.m function now includes new functionality including single and double bootstrap procedures, stratified resampling and calculation of bias-corrected parameter estimates, standard errors and confidence intervals
• changes were made to ibootnhst.m and maxstat.m to accomodate for changes in bootknife.m, and to enable displaying graph and output even when output arguments are called.

statistics-bootstrap-v3.6.10

• Improved coverage for double block bootstrap when block size is an odd number
• Added option to smooth bootknife sampling distribution

statistics-bootstrap-v3.6.9

• minor improvement to automatic block size calculation for block bootstrap (single and double bootstrap)