Main content

A Study on the Performance of the Aligned Rank Transform Procedure for Testing Interaction in Split-plot Designs, Using Count Data Generated from Neyman Type A Distributions

Show full item record

Title: A Study on the Performance of the Aligned Rank Transform Procedure for Testing Interaction in Split-plot Designs, Using Count Data Generated from Neyman Type A Distributions
Author: Yang, Wenjun
Department: Department of Mathematics and Statistics
Program: Mathematics and Statistics
Advisor: Balka, JeremyUmphrey, Gary
Abstract: Rank transformations are commonly employed in nonparametric alternatives to parametric procedures when standard distributional assumptions are violated. The Aligned Rank Transform (ART) procedure was developed to alleviate de ficiencies in simpler rank transform techniques when testing for interactions in factorial design ANOVAs. Most investigations have focussed on continuous data or regular factorial designs. This study compares the performance of the ART procedure to simpler ANOVA analyses on untransformed and square root transformed data from 3X3 CRD split-plot designs, where the data is simulated from Neyman Type A distributions. These distributions were derived to model clustered count data, such as insect counts. Parameter values were selected to include a distribution with a high proportion of zeroes and diverse clustering patterns for a fixed mean. The three methods showed similar robustness and power. For the range of parameters employed, the ART procedure did not demonstrate any advantage over simpler ANOVA procedures.
URI: http://hdl.handle.net/10214/14137
Date: 2018-08-30
Terms of Use: All items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.


Files in this item

Files Size Format View Description
Yang_Wenjun_201808_Msc.pdfuntranslated 2.994Mb PDF View/Open Finalized Thesis

This item appears in the following Collection(s)

Show full item record