Using SIMEX for smoothing-parameter choice in errors-in-variables problems

• 成果类型: Article
• 署名作者: Delaigle, Aurore; Hall, Peter
• 署名单位: University of Bristol; University of Melbourne
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1198/016214507000001355
• 发表日期: 2008
• 页码: 280-287
• 关键词: kernel density-estimation; simulation-extrapolation; Nonparametric Regression; bandwidth selection; Optimal Rates; deconvolution; models; CONVERGENCE; estimators

摘要

SIMEX methods are attractive for solving curve estimation problems in errors-in-variables regression, using parametric or semiparametric techniques. However, nonparametric approaches are generally of quite a different type, being based on, for example, kernels, local-linear modeling, ridging, orthogonal series, or splines. All of these techniques involve the challenging (and not well studied) issue of empirical smoothing parameter choice. We show that SIMEX can be used effectively for selecting smoothing parameters when applying nonparametric methods to errors-in-variable regression. In particular, we suggest an approach based on multiple error-inflated (or remeasured) data sets and extrapolation.