Published: June 17, 2017
favorite0 forum0 poll18
By: Andrew Bates, University of California San Diego
This was a project for a course on the bootstrap. I conducted a simulation study to investigate the performance of two alternatives to the bootstrap in situations where the bootstrap is known to fail. Emperical coverage probabilities of confidence intervals constructed using the m out of n bootstrap and subsampling were compared for several sample sizes and several resample sizes.