Investigating Bootstrap Confidence Intervals on Dr. D's 123

Document Type

Thesis

Publication Date

2008

Disciplines

Mathematics | Physical Sciences and Mathematics

Advisor

Michael Gass, Mathematics

Abstract

The focus of this thesis is to investigate the effect of varied selections of B, the number of Bootstrap samples, p, the probability associated with the discrete values, and n, the sample size, on the accuracy of confidence intervals developed by Biased-Corrected and Accelerated, and Percentile methods on Dr. D's 123. Efron, Tibshirani, and other statisticians have given general guidelines for selection the sample size and the number of Bootstrap samples associated with creating accurate confidence intervals. However, none of then particularly pertain to a distribution like Dr. D's 123. Our objective is to either validate their guidelines or form new ones pertaining to this distribution.

Share

COinS