Performs jackknife resampling on numpy arrays.
Jackknife resampling is a technique to generate ‘n’ deterministic samples of size ‘n-1’ from a measured sample of size ‘n’. Basically, the i-th sample, (1<=i<=n), is generated by means of removing the i-th measurement of the original sample. Like the bootstrap resampling, this statistical technique finds applications in estimating variance, bias, and confidence intervals.
data : numpy.ndarray
Original sample (1-D array) from which the jackknife resamples will be generated.
resamples : numpy.ndarray
The i-th row is the i-th jackknife sample, i.e., the original sample with the i-th measurement deleted.
[R52] McIntosh, Avery. “The Jackknife Estimation Method”. <http://people.bu.edu/aimcinto/jackknife.pdf> [R53] Efron, Bradley. “The Jackknife, the Bootstrap, and other Resampling Plans”. Technical Report No. 63, Division of Biostatistics, Stanford University, December, 1980. [R54] Cowles, Kate. “Computing in Statistics: The Jackknife, Lecture 11”. <http://homepage.stat.uiowa.edu/~kcowles/s166_2009/lect11.pdf>. September, 2009.