Bandwidth selection in robust smoothing
- 1 January 1993
- journal article
- research article
- Published by Taylor & Francis Ltd in Journal of Nonparametric Statistics
- Vol. 2 (4), 333-339
- https://doi.org/10.1080/10485259308832562
Abstract
In robust smoothing of regression functions of the type , the theory in automatic bandwidth selection is still lacking. This paper tries to fill the gap by looking at the use of cross-validation methods in robust smoothing. Conjectures regarding the use of a class of robust cross-validation method are made. These are further complemented by examples, all of which showed favourably for the new method.Keywords
This publication has 13 references indexed in Scilit:
- Adaptive $M$-Estimation in Nonparametric RegressionThe Annals of Statistics, 1990
- How Far Are Automatically Chosen Regression Smoothing Parameters From Their Optimum?Journal of the American Statistical Association, 1988
- Robust Nonparametric Regression with Simultaneous Scale Curve EstimationThe Annals of Statistics, 1988
- Optimal Bandwidth Selection in Nonparametric Regression Function EstimationThe Annals of Statistics, 1985
- Robust regression function estimationJournal of Multivariate Analysis, 1984
- How to determine the bandwidth of some nonlinear smoothers in practicePublished by Springer Science and Business Media LLC ,1984
- Smoothing noisy data with spline functionsNumerische Mathematik, 1978
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974
- Statistical predictor identificationAnnals of the Institute of Statistical Mathematics, 1970
- Robust Estimation of a Location ParameterThe Annals of Mathematical Statistics, 1964