February 7, 2001 Volume 29, No.14
|Title:||Modeling Long Distance Running Times by Age|
|Speaker:||Matthrew Strand, UCLA|
|Time:||Tuesday, February 13 at 4pm.|
|Title:||College Entrance and Cancer|
|Speaker:||Julie Legler, National Cancer Institute|
|Time:||Thursday, February 15 at 4pm.|
Two Statistics Talks
Two statisticians will be visiting us next week as candidates for a tenure track position in the Mathematics Department. All students are encouraged to come to the colloquia so that we can give our visitors a good sense of what the St. Olaf math department and its students are really like. Short biographies of Matt and Julie are on the back, as are the abstracts for their talks.
Matt has spent most of his life in Colorado. After growing up in the Denver area (Lakewood), he went to Colorado State University, where he obtained his B.S. degrees in math and statistics (89), and then an M.S. in
statistics (91). He then became a Peace Corps Volunteer, spending two years teaching math and science in a rural secondary school in Lesotho (southern Africa, 91-93). After this experience, he returned to Colorado, working with a research ecologist and taking secondary education courses at Colorado State. In the fall of 1995, Matt went back to school at Kansas State University for a Ph.D. in statistics, which he obtained in December of 1998. In the following year he was married to Linda Penuela (also of Colorado). Since the fall of 1999, Matt has worked as an assistant professor in the Department of Statistics at UCLA. In the summer of 2000, Matt and Linda had their first child, Evan Michael. In his spare time, Matt enjoys running and playing guitar.
Modeling Long Distance Running Times by Age
Matt's talk will be on his analysis of road
race times using regression and extreme value theory.
Matt's Research Talk
Matt will be giving a research level talk on Monday, February 12. The title is "Statistics and Sampling: Real-World Applications." The talk about how major organizations (Gallup, Bureau of Labor Statistics, Nielson, etc.) use sampling to obtain estimates for populations.
Julie Legler is currently employed by the National Cancer Institute in Rockville, Maryland. Her research involves the surveillance of cancer incidence and mortality trends. She earned her doctorate in biostatistics
from Harvard and a masters degree from the University of Minnesota in statistics. Special research interests include multiple binary endpoints, quality-of-life, and latent variable models.
College Entrance and Cancer
A statistical model referred to as an item response theory model is used in constructing standardized exams. This model employs a logistic function with a latent variable. It has been used to select items for college entrance exams. These models also provide ways to score tests as well. More recently, these models are finding application in the health sciences.
Julie's Research Talk
Julie will be giving a research level talk on Friday, February 16. The title is "Analyzing Multiple Binary Measures: a latent variable approach." At times, it is difficult to identify a single, primary endpoint for a study or clinical trial. Options are discussed for those cases when a set of multiple binary measures are available. The inner workings of the fitting
algorithm for a proposed model and some applications are presented.
To subscribe to the Math Mess, please contact Donna Brakke at email@example.com.
Editor-in-Chief: Jill Dietz
Associate Editor: Jennifer Beilfuss
Problems Editor: Cliff Corzatt
MM Czar: Donna Brakke