Mean and Location Estimation with Optimal Constants - UW Madison
Offered By: USC Probability and Statistics Seminar via YouTube
Course Description
Overview
Explore a comprehensive lecture on mean and location estimation in statistics, delivered by Jasper Lee from UW Madison. Delve into the fundamental challenges of accurately estimating distribution means from i.i.d. samples and determining location when the distribution shape is known. Examine the limitations of established asymptotic theories, including sample mean and maximum likelihood estimation, particularly in scenarios requiring high confidence with limited samples. Discover recent advancements in finite sample and high probability theories that yield optimal constants. Learn about new mean estimators for 1-dimensional and high-dimensional regimes with tight error bounds, as well as novel characterizations of location estimation that surpass traditional MLE approaches. Gain insights into future research directions in this field, based on collaborative work with Paul Valiant, Shivam Gupta, and Eric Price.
Syllabus
Jasper Lee: Mean and Location Estimation with Optimal Constants (UW Madison)
Taught by
USC Probability and Statistics Seminar
Related Courses
Statistical Inference and Modeling for High-throughput ExperimentsHarvard University via edX Estimation for Wireless Communications –MIMO/ OFDM Cellular and Sensor Networks
Indian Institute of Technology Kanpur via Swayam Обобщенные линейные модели
Saint Petersburg State University via Coursera Введение в теорию построения процедур множественной проверки гипотез
Higher School of Economics via Coursera Bayesian Statistics: Mixture Models
University of California, Santa Cruz via Coursera