Title:

The Shared Area Under Probability Curves Test: A Much-Less-Parametric Alternative to the Analysis of Variance

Video

Abstract

The commonly-used analysis of variance test has rigid assumptions. If these assumptions are not met, results can be misleading and erroneous. Non-parametric alternatives to the analysis of variance test use ranks that result in information (absolute distances between observations) being lost. The shared area under probability curves test is a new, non-parametric alternative to the analysis of variance that makes no assumptions about the data and that does not manipulate the data in any way that would cause information to be lost. This test calculates the amount of area under the curve shared by two probability distributions, each representing one group, to make inferences about whether groups are different.

Authors

First Name Last Name
David Moore

File Count: 1


Leave a comment

Comments are viewable only by submitter



Submission Details

Conference GRC
Event Graduate Research Conference
Department Natural Resources and Earth Systems Science (GRC)
Group Oral Presentation
Added April 16, 2020, 11:51 a.m.
Updated April 16, 2020, 4:20 p.m.
See More Department Presentations Here