Seminar at SMU Delhi
March 25, 2015 (Wednesday) ,
3:30 PM at Webinar
Speaker:
Hassan Doosti,
Mashhad University of Medical Sciences, Mashhad, Iran
Title:
Making a Nonparametric Density Estimator More Attractive, and More Accurate, by Data Perturbation
Abstract of Talk
Motivated by both the shortcomings of high-order density estimators, and the
increasingly large datasets in many areas of modern science, in this talk, we
introduce new high-order, nonparametric density estimators that are guaranteed
to be positive and do not have highly oscillatory tails. Our approach is based
on data perturbation, for example by tilting or data sharpening. It leads to
new estimators that are more accurate than conventional kernel techniques that
use positive kernels, but which nevertheless enjoy the positivity property, and
are far less ``wiggly'' than high-order kernel estimators. We investigate
performance by theoretical~analysis and in a numerical study. [Joint work with Peter Hall: To appear in JRSS-B].