Advances and Applications in Statistics
Volume 48, Issue 2, Pages 109 - 121
(February 2016) http://dx.doi.org/10.17654/AS048020109 |
|
AN APPROACH FOR BIAS REDUCTION IN DENSITY ESTIMATION
Kee-Hoon Kang
|
Abstract: Bias reduction in nonparametric function estimation is an important issue and enough work has been done on this topic. In this paper, we consider the data sharpening approach, which is a useful tool for enhancing statistical properties of conventional nonparametric curve estimators. We propose a simple numerical algorithm for data sharpening in density estimation to reduce biases. It does not require the estimation procedures of density and its derivatives. Numerical results reveal that our algorithm works well and is computationally efficient. |
Keywords and phrases: bandwidth, data sharpening, EM algorithm. |
|
Number of Downloads: 445 | Number of Views: 1352 |
|