Image

Bing Li, Verne M. Willaman Professor of Statistics, will present a talk titled “Sufficiency: A Brief History of Data Reduction." His talk will explore the ideas of data reduction—simplifying data without losing important information—and of data sufficiency—comprehensively summarizing data using relatively few statistics. Modern data is often complex and has high dimension, meaning more features are considered (e.g., height, weight, and genetic information in a medical study) than observations (e.g., individual patients). Statisticians develop methods to reduce this dimensionality and make analysis and interpretation easier. Efficient methods of dimension reduction are useful in a variety of applications, such as pattern recognition, classification, statistical learning, medical research, and bioinformatics.

In this talk, Li will give an overview of the evolution of sufficiency in data reduction and explain how it has permeated many aspects of contemporary statistics. He will focus on two areas of his research where sufficiency has played a key role: sufficient dimension reduction and sufficient graphical models.

A reception will follow the talk in the Verne M. Willaman Gateway to the Sciences on the third-floor bridge connecting the Huck Life Sciences and Chemistry Buildings.

To attend the lecture via Zoom, please visit this Zoom link.