On bandwidth choice for density estimation with dependent data
Open Access
- 1 December 1995
- journal article
- Published by Institute of Mathematical Statistics in The Annals of Statistics
- Vol. 23 (6), 2241-2263
- https://doi.org/10.1214/aos/1034713655
Abstract
We address the empirical bandwidth choice problem in cases where the range of dependence may be virtually arbitrarily long. Assuming that the observed data derive from an unknown function of a Gaussian process, it is argued that, unlike more traditional contexts of statistical inference, in density estimation there is no clear role for the classical distinction between short- and long-range dependence. Indeed, the "boundaries" that separate different modes of behaviour for optimal bandwidths and mean squared errors are determined more by kernel order than by traditional notions of strength of dependence, for example, by whether or not the sum of the covariances converges. We provide surprising evidence that, even for some strongly dependent data sequences, the asymptotically optimal bandwidth for independent data is a good choice. A plug-in empirical bandwidth selector based on this observation is suggested. We determine the properties of this choice for a wide range of different strengths of dependence. Properties of cross-validation are also addressed.Keywords
This publication has 34 references indexed in Scilit:
- Density Estimation in the $L^\infty$ Norm for Dependent Data with Applications to the Gibbs SamplerThe Annals of Statistics, 1993
- Asymptotic behaviour of the mean integrated squared error of kernel density estimators for dependent observationsThe Canadian Journal of Statistics / La Revue Canadienne de Statistique, 1990
- Data-Driven Bandwidth Choice for Density Estimation Based on Dependent DataThe Annals of Statistics, 1990
- Convergence rates in density estimation for data from infinite-order moving average processesProbability Theory and Related Fields, 1990
- Asymptotic normality of the kernel estimate under dependence conditions: application to hazard rateJournal of Statistical Planning and Inference, 1990
- The L/sub 1/ and L/sub 2/ strong consistency of recursive kernel density estimation from dependent samplesIEEE Transactions on Information Theory, 1990
- Comparison of Data-Driven Bandwidth SelectorsJournal of the American Statistical Association, 1990
- How Far Are Automatically Chosen Regression Smoothing Parameters From Their Optimum?: CommentJournal of the American Statistical Association, 1988
- Nonparametric Density Estimation, Prediction, and Regression for Markov SequencesJournal of the American Statistical Association, 1985
- Density Estimation in a Continuous-Time Stationary Markov ProcessThe Annals of Statistics, 1979