Absolute error criteria for bandwidth selection in density estimation from censored data

J. W. Kuhn, W. J. Padgett, J. G. Surles

Research output: Contribution to journalArticlepeer-review

Abstract

In kernel density estimation, a criticism of bandwidth selection techniques which minimize squared error expressions is that they perform poorly when estimating tails of probability density functions. Techniques minimizing absolute error expressions are thought to result in more uniform performance and be potentially superior. An asymptotic mean absolute error expression for nonparametric kernel density estimators from right-censored data is developed here. This expression is used to obtain local and global bandwidths that are optimal in the sense that they minimize asymptotic mean absolute error and integrated asymptotic mean absolute error, respectively. These estimators are illustrated for eight data sets from known distributions. Computer simulation results are discussed, comparing the estimation methods with squared-error-based bandwidth selection for right-censored data.

Original languageEnglish
Pages (from-to)215-230
Number of pages16
JournalJournal of Statistical Computation and Simulation
Volume70
Issue number3
DOIs
StatePublished - 2001

Keywords

  • Asymptotic mean absolute error
  • Bandwidth selection
  • Lifetime data
  • Nonparametric density estimation
  • Random censorship
  • Smoothing parameter

Fingerprint Dive into the research topics of 'Absolute error criteria for bandwidth selection in density estimation from censored data'. Together they form a unique fingerprint.

Cite this