A technique for establishing L1 asymptotic unbiasedness of a kernel density
estimator in Rd that does not depend on the form of the kernel function will be
demonstrated. We will introduce the concept of a region sequence of a sequence
of kernel functions and show how this can be used to give necessary and sufficient
conditions for L1 asymptotic unbiasedness. These results are then applied to kernel
density estimators whose form is given and a number of known and novel results
are obtained.
| Identifer | oai:union.ndltd.org:MANITOBA/oai:mspace.lib.umanitoba.ca:1993/22110 |
| Date | 26 August 2013 |
| Creators | Stinner, Mark |
| Contributors | Leblanc, Alexandre (Statistics), Koulis, Theodoro (Statistics) Prymak, Andriy (Mathematics) |
| Source Sets | University of Manitoba Canada |
| Detected Language | English |
Page generated in 0.0015 seconds