Online Program Home
My Program

Abstract Details

Activity Number: 523 - Contributed Poster Presentations: ENAR
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 10:30 AM to 12:20 PM
Sponsor: ENAR
Abstract #307186
Title: Simultaneous Estimations for Contrasts of Quantiles
Author(s): Gemechis Djira* and Lawrence Segbehoe and Frank Schaarschmidt
Companies: South Dakota State University and South Dakota State University and Leibniz Universit├Ąt Hannover
Keywords: Asymptotic; Simultaneous inference; Skewed distribution; Ratios; Quantiles

Skewed data are ubiquitous in practice. Although the mean is popular, many researches in the health and social sciences involve skewed distributions and inferences about quantiles. Many of the existing simultaneous inference procedures are for the mean which is sensitive to skewness or outlying observations. We develop an asymptotic method for constructing simultaneous confidence intervals for multiple contrasts of quantiles associated with independent and arbitrary distributions. This includes, for example, comparisons with a control using medians. Small-sample performance of the proposed method is assessed in terms of simultaneous coverage probability and average width of the confidence intervals. Good coverage probabilities are observed even for extremely skewed distributions. The proposed method is applied to biomedical data following a log-normal distribution and time-to-event data in survival analysis.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program