SFB 303 Discussion Paper No. A - 112
Author: Härdle, Wolfgang
Title: Sequential Kernel Smoothing for Estimation of Zeros and Location of Extrema of Regression Functions
Abstract: An enumeration of zeros and of locations of extrema often suffices to describe the approximate shape of
regression functions. The estimation of these quantities is not only quantifying the shape of the functions but also
offers the possibility of comparing shapes in a group of similar shaped functions. An illustrating example is the
human height growth curve, where location and size of a peak in the second derivative of this regression curve
serve to describe the so-called midgrowth spurt, see Müller (1985). Such "longitudinal parameters", as they have
been called by Gasser et al. (1984), can be used for group comparisons.
Creation-Date: May 1987
Unfortunately this paper is not available online. Please contact us to order a hardcopy.
SFB 303 Homepage