Bootstrap Confidence Interval of Prediction for Small Area Estimation Based on Linear Mixed Model

F. Novkaniza, K. A. Notodiputro, B. Sartono

Research output: Contribution to journalConference articlepeer-review

1 Citation (Scopus)

Abstract

Linear Mixed Model (LMM) analyzes the relationship between Gaussian response and predictors with either fixed and random effects. Procedures based on LMM have been used to construct estimates of the means of small areas, by exploiting auxiliary information. In this article, we show how to resample fixed effects coefficient estimates via bootstrapping and we construct nonparametric and parametric bootstrap confidence interval of predictions for small area estimation, based on mixed-effects linear models. Examples of computation for bootstrap confidence intervals of prediction are given for Battese, Harter and Fuller Data (1988).

Original languageEnglish
Article number012040
JournalIOP Conference Series: Earth and Environmental Science
Volume187
Issue number1
DOIs
Publication statusPublished - 19 Nov 2018
Event4th International Seminar on Sciences, ISS 2017 - Baranangsiang, Bogor, Indonesia
Duration: 19 Oct 201720 Oct 2017

Keywords

  • mixed model
  • nonparametric bootstrap
  • parametric bootstrap
  • prediction interval

Fingerprint

Dive into the research topics of 'Bootstrap Confidence Interval of Prediction for Small Area Estimation Based on Linear Mixed Model'. Together they form a unique fingerprint.

Cite this