PLACE:  319 Snedecor

SPEAKER:
Dan Nettleton
Iowa State University

TITLE:
Testing the Equality of Distributions of Random Vectors with Categorical Components

ABSTRACT:

We develop a method for testing the equality of two or more distributions of random vectors with categorical components.  A complete graph of the data is constructed using the observed data vectors as nodes.The union of edge sets from all nearest-neighbor subgraphs of the complete graph is determined.  The test statistic is the number of edges in this union that link observations from different distributions.  Inference is conditional on the number of observations from each distribution and the number of times each of the data vectors is observed in the pooled sample.  Permutation testing and asymptotics are used to estimate the observed significance level.  Analysis of data sets from medicine, boxing,
and an introductory statistics course will be presented.  No knowledge of graph theory is needed to understand the talk.

COFFEE: 3:40 p.m., 104 Snedecor