Year in Review: Undergraduate Statistics Program
During the August 2017 commencement ceremonies, Texas A&M University awarded diplomas to the largest summer class in its 140-year history — a group that included the first two graduates of one of its newest degree programs, the bachelor of science in statistics. Texas A&M statistician Alan Dabney, one of two faculty advisors for the program, summarized his thoughts on the program’s historic first year — 12 months that helped establish a firm foundation for both the students enrolled and the Department of Statistics, as well as within a broader profession with the potential to impact so many others.
See the full article in the August edition of Science Communications' discover-e publication.
Texas A&M - Labors of Lab
A new "Labors of Lab" episode on statistics Ph.D candidate Raanju Sundararajan, Class of 2017. Studying under the advisement of Prof. Mohsen Pourahmadi, Raanju discusses his research in time series analysis, which he's using to model earthquake occurrence patterns, as well as one of his greatest learning opportunities thus far at Texas A&M: teaching.
Texas A&M Statistician Cliff Spiegelman to Serve as Expert Witness in State of Texas v. Lee Harvey Oswald Mock TrialTexas A&M statistician Cliff Spiegelman is one of a handful of world-renowned John F. Kennedy assassination experts offering their pro bono services for a two-day mock trial, State of Texas v. Lee Harvey Oswald, set to play out November 16 and 17 on the campus of Houston's oldest law school, South Texas College of Law Houston.Read More →
The sky's the limit for astronomy, data-driven science and Texas A&M statistician Jianhua Huang, who has been appointed as the inaugural holder of the Arseven/Mitchell Chair in Astronomical Statistics intended to accelerate the development of related capabilities at Texas A&M.Read More →
11:30 AM / 12:30 PM Blocker Building (BLOC), Room 113 979-845-3141
Department of Data Science and Operations
University of Southern California
Graph-Guided Banding for Covariance Estimation
Reliable estimation of the covariance matrix is notoriously difficult in high dimensions. Numerous methods assume that the population covariance (or inverse covariance) matrix is sparse while making no particular structural assumptions on the desired sparsity pattern. A highly-related, yet complementary, literature studies the setting in which the measured variables have a known ordering, in which case a banded (or near-banded) population matrix is assumed. This talk will address the broad middle ground that lies between the former approach of complete neutrality to the sparsity pattern and the latter highly restrictive assumption of having a known ordering. A class of convex regularizers is developed that is in the spirit of banding and yet attains sparsity structures that can be customized to a wide variety of applications.
Friday, 11/17/2017, BLOC 113, 11:30 AM
12:30 PM / 01:30 AM Blocker Building (BLOC), Room 521 979-845-3141
Ph.D. Student, Department of Statistics
Texas A&M University
Categorizing a Continuous Predictor Subject to Measurement
Epidemiologists often categorize a continuous risk predictor, even when the true risk model is not a categorical one. Nonetheless, such categorization is thought to be more interpretable, and thus their goal is to fit the categorical model and interpret the categorical parameters. We address the question: with measurement error and categorization, how can we do what epidemiologists want, namely to estimate the parameters of the categorical model that would have been estimated if the true predictor was observed? We develop a general methodology for such an analysis, and illustrate it in linear and logistic regression. Simulation studies are presented and the methodology is applied to a real data set.
Monday, 11/20/17, 12:30 PM, BLOC 521