On some predictors of times to failure of censored items in progressively censored samples

Indrani Basak, Prasanta Basak, N. Balakrishnan

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

In this article, we consider the problem of predicting times to failure of units censored in multiple stages in a progressively censored sample from an absolutely continuous population. The best linear unbiased predictors (BLUP), the maximum-likelihood predictors (MLP), and the conditional median predictors (CMP) are considered. The properties of MLP such as unbiasedness, consistency and efficiency are examined. The MLP or modified MLP (MMLP) are derived for exponential and extreme value populations. In addition, for these populations, the conditional distributions are used to derive the CMP. Comparison of different predictors are made with respect to mean squared prediction error (MSPE). Finally, some numerical examples are presented to illustrate all the prediction methods discussed here. Using simulation studies, prediction intervals are also generated for these examples.

Original languageEnglish (US)
Pages (from-to)1313-1337
Number of pages25
JournalComputational Statistics and Data Analysis
Volume50
Issue number5
DOIs
StatePublished - Mar 1 2006

Fingerprint

Censored Samples
Maximum likelihood
Predictors
Maximum Likelihood
Best Linear Unbiased Predictor
Unbiasedness
Prediction Interval
Extreme Values
Prediction Error
Absolutely Continuous
Conditional Distribution
Mean Squared Error
Simulation Study
Numerical Examples
Unit
Prediction

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

@article{cb4ad4868b56472fb3ea49a4e463231f,
title = "On some predictors of times to failure of censored items in progressively censored samples",
abstract = "In this article, we consider the problem of predicting times to failure of units censored in multiple stages in a progressively censored sample from an absolutely continuous population. The best linear unbiased predictors (BLUP), the maximum-likelihood predictors (MLP), and the conditional median predictors (CMP) are considered. The properties of MLP such as unbiasedness, consistency and efficiency are examined. The MLP or modified MLP (MMLP) are derived for exponential and extreme value populations. In addition, for these populations, the conditional distributions are used to derive the CMP. Comparison of different predictors are made with respect to mean squared prediction error (MSPE). Finally, some numerical examples are presented to illustrate all the prediction methods discussed here. Using simulation studies, prediction intervals are also generated for these examples.",
author = "Indrani Basak and Prasanta Basak and N. Balakrishnan",
year = "2006",
month = "3",
day = "1",
doi = "10.1016/j.csda.2005.01.011",
language = "English (US)",
volume = "50",
pages = "1313--1337",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "5",

}

On some predictors of times to failure of censored items in progressively censored samples. / Basak, Indrani; Basak, Prasanta; Balakrishnan, N.

In: Computational Statistics and Data Analysis, Vol. 50, No. 5, 01.03.2006, p. 1313-1337.

Research output: Contribution to journalArticle

TY - JOUR

T1 - On some predictors of times to failure of censored items in progressively censored samples

AU - Basak, Indrani

AU - Basak, Prasanta

AU - Balakrishnan, N.

PY - 2006/3/1

Y1 - 2006/3/1

N2 - In this article, we consider the problem of predicting times to failure of units censored in multiple stages in a progressively censored sample from an absolutely continuous population. The best linear unbiased predictors (BLUP), the maximum-likelihood predictors (MLP), and the conditional median predictors (CMP) are considered. The properties of MLP such as unbiasedness, consistency and efficiency are examined. The MLP or modified MLP (MMLP) are derived for exponential and extreme value populations. In addition, for these populations, the conditional distributions are used to derive the CMP. Comparison of different predictors are made with respect to mean squared prediction error (MSPE). Finally, some numerical examples are presented to illustrate all the prediction methods discussed here. Using simulation studies, prediction intervals are also generated for these examples.

AB - In this article, we consider the problem of predicting times to failure of units censored in multiple stages in a progressively censored sample from an absolutely continuous population. The best linear unbiased predictors (BLUP), the maximum-likelihood predictors (MLP), and the conditional median predictors (CMP) are considered. The properties of MLP such as unbiasedness, consistency and efficiency are examined. The MLP or modified MLP (MMLP) are derived for exponential and extreme value populations. In addition, for these populations, the conditional distributions are used to derive the CMP. Comparison of different predictors are made with respect to mean squared prediction error (MSPE). Finally, some numerical examples are presented to illustrate all the prediction methods discussed here. Using simulation studies, prediction intervals are also generated for these examples.

UR - http://www.scopus.com/inward/record.url?scp=27644531117&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27644531117&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2005.01.011

DO - 10.1016/j.csda.2005.01.011

M3 - Article

VL - 50

SP - 1313

EP - 1337

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 5

ER -