Publication details

Jádrové odhady regresní funkce

Investor logo
Title in English Kernel Estimation of the Regression Function


Year of publication 2005
MU Faculty or unit

Faculty of Science

Description The problem of bandwidth selection for non--parametric kernel regression is considered. We follow the local polynomial estimators and present some their statistical properties. We also demonstrate the effect of the smoothing parameter on the quality of the smoothed curve and introduce some classical methods for bandwidth selection.\\ The cyclic design is assumed in this work to avoid the difficulties caused by boundary effects. Most of bandwidth selectors are based on the residual sum of squares (RSS). It is often observed in simulation studies that these selectors are biased toward to undersmoothing. This leads to consideration of a procedure which stabilizes the RSS by modifying the periodogram of the observations.\\ Simulation studies suggest that the proposed selector is much more consistent than the~classical one. But in practical examples, there the results can be miscellaneous.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.

More info