Improved Asymptotics for Zeros of Kernel Estimates via a Reformulation of the LeadbetterCryer Integral
Abstract
The expected number of false inflection points of kernel smoothers is evaluated. To obtain the small noise limit, we use a reformulation of the LeadbetterCryer integral for the expected number of zero crossings of a differentiable Gaussian process.
 Publication:

arXiv eprints
 Pub Date:
 March 2018
 arXiv:
 arXiv:1803.03896
 Bibcode:
 2018arXiv180303896R
 Keywords:

 Statistics  Methodology;
 Electrical Engineering and Systems Science  Signal Processing;
 Mathematics  Probability;
 Mathematics  Statistics Theory;
 Physics  Data Analysis;
 Statistics and Probability
 EPrint:
 Statistics &