6.2020 - Kernel smoothing methods, theory and practice

Intership in statistics

Laurent DELSOL, University of Orleans

Theoretical Background: Probabilities, introduction to mathematical statistics
Tools: R software

Kernel smoothing methods, theory and practice
The aim of this internship is to go further parametric estimators and consider kernel
estimators designed for non parametric curve (e.g. density or regression functions)
estimation. Both theoretical and practical aspects will be considered. The first step
will be to define the estimators, understand the role of each parameter and prove
some asymptotic results. Clustering and supervised classification are potential
interesting applications. The implementation of the considered methods will be done
on R software.

Bibliographical references :
Kernel Smoothing M.P. Wand, M.C. Jones CRC Press, 1 déc. 1994 - 224 pages