Dr. Brady T. West will be giving two talks. One will be at 3:30pm (see here) and the other will begin at 4:30pm (see here)

This award for excellence in teaching annually honors outstanding faculty in the Franklin College who have shown a sustained commitment to high quality instruction. Abhyuday is a very popular and successful instructor, as well as an outstanding mentor for students interested in undergraduate research or further study. He is a highly regarded supervisor of graduate student research and a valuable contributor to teaching-related outreach.

Agenda

3:30 - 4:00pm- Arrival 

4:00 - 5:00pm- Lecture, Dr. Art Owen, Stanford University

5:00 - 6:00pm- Happy Hour (light refreshments will be served)

6:00 - 8:00pm- Dinner and After-Dinner Talk, Dr. Art Owen, Stanford University

Takashi Owada is an assistant professor of Department of Statistics in Purdue University. Prior to joining Purdue, he was a postdoc researcher at Technion-Israel Institute of Technology, and received his PhD in Operations Research at Cornell University. His research interests lie in random topology, topological data analysis, random graph theory, heavy tail probability, extreme value theory, and long range dependence. 

Robust Estimation under Huber's Contamination Model

This talk describes some new challenges and results in high-dimensional and nonparametric statistics under the celebrated Huber’s contamination model. We particularly focus on the influence of contamination on the minimax rates and the corresponding rate-optimal procedures.

Mediation Analysis of High-Dimensional Microbiome and Host Genome Data

Estimating diversity and relative abundance in microbial communities

Nonparametric Estimation of a Two-component Mixture Model with application to Multiple Testing   

We consider estimation and inference in a two-component mixture model where the distribution of one component is completely unknown. We develop methods for estimating the mixing proportion and

Assessing Procedures vs Assessing Evidence

Many statistical analyses are characterized by how often a procedure works: how often an interval covers a true value, a null hypothesis is rejected, an item is correctly classified, etc. But assessing how often a procedure works differs from assessing the evidence in a data set. Understanding the difference is prerequisite to understanding what matters in a given analysis: the procedure, the evidence, or both.