A comparative study of four estimators for analyzing the random event rate of the poissan process

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

In this paper, a random effect Poissan regression model is considered for the prediction of the failure rate which would follow a lognormal distribution. A two stage procedure is used to obtain the regression estimator of the failure rate as well as the shrinkage estimator. These estimators are compared to both the raw estimator which entirely depends on the historical failure records and a shrinkage estimator in which a gamma distribution is used mistakenly in place of the lognormal prior distribution. Results of Monte-Carlo simulation indicate the following in terms of the MSE: (1) overall, the shrinkage estimator based on the lognormal prior distribution performs best; (2) with the moderate variability in the failure rates (0-2.5), the performance of the shrinkage estimator based on the gamma distribution is not significantly different from that of the shrinkage estimator based on the lognormal distribution; (3) when there exists considerable variability in the failure rates (0–10), the raw estimator appears to replace shrinkage estimators. In terms of the Bias, the raw estimator performs better than the others.

Original languageEnglish
Pages (from-to)1-10
Number of pages10
JournalJournal of Statistical Computation and Simulation
Volume49
Issue number1-2
DOIs
Publication statusPublished - 1994 Jan 1

Fingerprint

Shrinkage Estimator
Comparative Study
Log Normal Distribution
Failure Rate
Estimator
Gamma distribution
Prior distribution
Two-stage Procedure
Regression Estimator
Random Effects Model
Shrinkage estimator
Comparative study
Regression Model
Monte Carlo Simulation
Failure rate
Prediction

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modelling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

@article{7821c84b9280416aa32df9d3043604ee,
title = "A comparative study of four estimators for analyzing the random event rate of the poissan process",
abstract = "In this paper, a random effect Poissan regression model is considered for the prediction of the failure rate which would follow a lognormal distribution. A two stage procedure is used to obtain the regression estimator of the failure rate as well as the shrinkage estimator. These estimators are compared to both the raw estimator which entirely depends on the historical failure records and a shrinkage estimator in which a gamma distribution is used mistakenly in place of the lognormal prior distribution. Results of Monte-Carlo simulation indicate the following in terms of the MSE: (1) overall, the shrinkage estimator based on the lognormal prior distribution performs best; (2) with the moderate variability in the failure rates (0-2.5), the performance of the shrinkage estimator based on the gamma distribution is not significantly different from that of the shrinkage estimator based on the lognormal distribution; (3) when there exists considerable variability in the failure rates (0–10), the raw estimator appears to replace shrinkage estimators. In terms of the Bias, the raw estimator performs better than the others.",
author = "Sohn, {So Young}",
year = "1994",
month = "1",
day = "1",
doi = "10.1080/00949659408811556",
language = "English",
volume = "49",
pages = "1--10",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "1-2",

}

TY - JOUR

T1 - A comparative study of four estimators for analyzing the random event rate of the poissan process

AU - Sohn, So Young

PY - 1994/1/1

Y1 - 1994/1/1

N2 - In this paper, a random effect Poissan regression model is considered for the prediction of the failure rate which would follow a lognormal distribution. A two stage procedure is used to obtain the regression estimator of the failure rate as well as the shrinkage estimator. These estimators are compared to both the raw estimator which entirely depends on the historical failure records and a shrinkage estimator in which a gamma distribution is used mistakenly in place of the lognormal prior distribution. Results of Monte-Carlo simulation indicate the following in terms of the MSE: (1) overall, the shrinkage estimator based on the lognormal prior distribution performs best; (2) with the moderate variability in the failure rates (0-2.5), the performance of the shrinkage estimator based on the gamma distribution is not significantly different from that of the shrinkage estimator based on the lognormal distribution; (3) when there exists considerable variability in the failure rates (0–10), the raw estimator appears to replace shrinkage estimators. In terms of the Bias, the raw estimator performs better than the others.

AB - In this paper, a random effect Poissan regression model is considered for the prediction of the failure rate which would follow a lognormal distribution. A two stage procedure is used to obtain the regression estimator of the failure rate as well as the shrinkage estimator. These estimators are compared to both the raw estimator which entirely depends on the historical failure records and a shrinkage estimator in which a gamma distribution is used mistakenly in place of the lognormal prior distribution. Results of Monte-Carlo simulation indicate the following in terms of the MSE: (1) overall, the shrinkage estimator based on the lognormal prior distribution performs best; (2) with the moderate variability in the failure rates (0-2.5), the performance of the shrinkage estimator based on the gamma distribution is not significantly different from that of the shrinkage estimator based on the lognormal distribution; (3) when there exists considerable variability in the failure rates (0–10), the raw estimator appears to replace shrinkage estimators. In terms of the Bias, the raw estimator performs better than the others.

UR - http://www.scopus.com/inward/record.url?scp=84972812789&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84972812789&partnerID=8YFLogxK

U2 - 10.1080/00949659408811556

DO - 10.1080/00949659408811556

M3 - Article

VL - 49

SP - 1

EP - 10

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 1-2

ER -