Author ORCID Identifier

https://orcid.org/0009-0009-6097-5874

Date of Award

8-7-2024

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Mathematics and Statistics

First Advisor

Yichuan Zhao

Abstract

Chatterjee’s rank correlation coefficient (CCC) is a rank correlation coefficient that estimates a population quantity while remaining asymptotically normal under independence. However, it has been proven to be rate suboptimal compared to other recognized rank correlations. CCC has been demonstrated to adapt to the data’s complex structure, making it a potentially useful tool. This thesis applies the jackknife empirical likelihood (JEL) and its variation to estimate confidence intervals for CCC. It compares coverage probability and interval length for JEL and normal approximation (NA), bootstrap, and adjusted jackknife empirical likelihood (AJEL) methods. Simulation studies are carried out to assess the suggested estimators’ performance. Simulation results showed that the AJEL methods having the widest confidence intervals in most cases perform better than the NA methods in terms of coverage rate of the confidence interval. Also, bootstrap showed inconsistent coverage probability. The performances of the proposed methods are tested using real data analysis.

DOI

https://doi.org/10.57709/37442124

File Upload Confirmation

1

Available for download on Sunday, August 02, 2026

Share

COinS