|
Activity Number:
|
335
- SPEED: Reliable Statistical Learning and Data Science
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
|
Abstract #322510
|
View Presentation
|
|
Title:
|
Linear Combinations of Percentiles for Tests in the One Way Layout
|
|
Author(s):
|
Yvonne Zubovic*
|
|
Companies:
|
Indiana University Purdue University Fort Wayne
|
|
Keywords:
|
Percentiles ;
one-way layout ;
power
|
|
Abstract:
|
We consider the case when the sample size and several sample percentiles are the only available information from k normally distributed populations. We discuss a statistic for testing the equality of the k population means based on a linear combination of three or more available percentiles from each sample. The work considers the cases when sample sizes are unequal. Also considered is the case when different percentiles are available across samples. Under the assumption of normality, the distribution of the proposed test statistic can be approximated using an F distribution. Simulation results are presented for the Type I error rate and power of the proposed test.
|
Authors who are presenting talks have a * after their name.
