Date of Award
8-2009
Document Type
Thesis
Degree Name
Master of Science (MS)
Legacy Department
Mathematical Science
Committee Chair/Advisor
Kulasekera, Karunarathna B
Committee Member
Park , Chanseok
Abstract
We consider nonparametric estimation of a smooth regression function of one variable. In practice it is quite popular to use the data to select one global smoothing parameter. Such global selection procedures cannot sufficiently account for local sparseness of the covariate nor can they adapt to local curvature of the regression function. We propose a new method to select local smoothing parameters which takes into account sparseness and adapts to local curvature of the regression function. A Bayesian method allows the smoothing parameter to adapt to the local sparseness of the covariate and provides the basis for a local cross validation procedure which adjusts smoothing according to local curvature of the regression function. Simulation evidence indicates that the method can result in significant reduction of both point-wise mean squared error and integrated mean squared error of the estimators.
Recommended Citation
Zheng, Qi, "Local Adaptive Smoothing in Kernel Regression Estimation" (2009). All Theses. 616.
https://tigerprints.clemson.edu/all_theses/616