Some lessons on Generalized Linear Models: the horseshoe crabs data"


September 14, 2010
A presentation by Rob Kushler


A data set containing measured characteristics of a sample of horseshoe crabs is used extensively in the text on discrete data methods by Agresti (Wiley 2002). In this talk several issues involved in fitting Generalized Linear Models (GLMs) are illustrated using this data set, and a previously unrecognized anomaly is revealed and resolved.

Robert Kushler is an Associate Professor in the Department of Mathematics and Statistics at Oakland University. He has an M.A. in Statistics and a Ph.D. in Biostatistics, both from the University of Michigan. His research interests include development of new methodology for statistical modeling,
computer experiments, and industrial applications of statistics. He has extensive consulting experience in the automotive industry, and also provides statistical consulting to students and faculty on campus at Oakland University. He is the current Treasurer of the Detroit Chapter.

 

Scott Millis and George Divine giving Rob Kushler a Certificate of Appreciation from the Detroit Chapter