Taylor & Francis Group
Browse
gnst_a_1770754_sm1405.pdf (215.96 kB)

Higher-order bias corrections for kernel type density estimators on the unit or semi-infinite interval

Download (215.96 kB)
journal contribution
posted on 2020-05-27, 11:50 authored by Gaku Igarashi, Yoshihide Kakizawa

For the data of size n from the unit or semi-infinite interval, several asymmetric kernel density estimators, having the mean integrated squared errors of order O(n4/5) or O(n8/9), are available in the literature. We develop more higher-order bias-corrected estimators, achieving the order O(n4p/(4p+1)), where p2 is a given integer. We illustrate the finite sample performance of the estimators through the simulations.

Funding

The first author has been supported by the Japan Society for the Promotion of Science (JSPS); Grant-in-Aid for Young Scientists (B), 17K13714. The second author has been supported by the JSPS; Grant-in-Aid for Scientific Research (C), 17K00041.

History