Some Errors Associated with the Non-parametric Estimation of Density Functions
- 1 December 1976
- journal article
- research article
- Published by Oxford University Press (OUP) in IMA Journal of Applied Mathematics
- Vol. 18 (3) , 371-380
- https://doi.org/10.1093/imamat/18.3.371
Abstract
Explicit expressions are obtained for the bias, mean square, expected mean square and integrated mean square errors associated with the estimation of the normal probability density function using a Gaussian kernel. It is shown that this estimator is relatively robust to departures from normality and hence leads to satisfactory estimates under quite wide conditions. Two methods of bias reduction are suggested.Keywords
This publication has 0 references indexed in Scilit: