Authors: Dette, Holger
Weißbach, Rafael
Title: Bias in nearest-neighbor hazard estimation
Language (ISO): en
Abstract: In nonparametric curve estimation, the smoothing parameter is critical for performance. In order to estimate the hazard rate, we compare nearest neighbor selectors that minimize the quadratic, the Kullback-Leibler, and the uniform loss. These measures result in a rule of thumb, a crossvalidation, and a plug-in selector. A Monte Carlo simulation within the threeparameter exponentiated Weibull distribution indicates that a counterfactual normal distribution, as an input to the selector, does provide a good rule of thumb. If bias is the main concern, minimizing the uniform loss yields the best results, but at the cost of very high variability. Crossvalidation has a similar bias to the rule of thumb, but also with high variability. AMS: 62M02
Subject Headings: Bandwidth selection
Hazard rate
Kernel estimation
Nearest neighbor bandwidth
Rule of thumb
Variable bandwidth
Issue Date: 2008-11-26T14:48:51Z
Appears in Collections:Sonderforschungsbereich (SFB) 475

Files in This Item:
File Description SizeFormat 
tr15-08-Weißbach.pdfDNB371 kBAdobe PDFView/Open

This item is protected by original copyright

Items in Eldorado are protected by copyright, with all rights reserved, unless otherwise indicated.