Department of Statistics Seminar
North Carolina State University
presents
Dr. Nancy Glenn
University of South Carolina
"Weighted Empirical Likelihood"
ABSTRACT
Weighted empirical likelihood is a nonparametric likelihood function that is less sensitive to outliers than the empirical likelihood function. The technique iteratively weights data based on scale estimation, then employs constrained optimization to derive point estimators. The iterative reweighting process allows accurate parameter estimation despite any potential data contamination.
Friday, April, 23, 2004
3:35 - 4:35 pm
206 Cox Hall
Refreshments will be served on the second floor of Dabney Hall (left of Room 222) at 3:00 pm.