Methods for comparing the means of two independent log-normal samples

Xiao Hua Zhou, Sujuan Gao, Siu Hui

Research output: Contribution to journalArticle

113 Citations (Scopus)

Abstract

Standard methods of using the t-test and the Wilcoxon test have deficiencies for comparing the means of two skewed log-normal samples. In this paper, we propose two new methods to overcome these deficiencies: (1) a likelihood-based approach and (2) a bootstrap-based approach. Our simulation study shows that the likelihood-based approach is the best in terms of the type I error rate and power when data follow a log-normal distribution.

Original languageEnglish
Pages (from-to)1129-1135
Number of pages7
JournalBiometrics
Volume53
Issue number3
DOIs
StatePublished - Sep 1997

Fingerprint

Normal distribution
Likelihood
Wilcoxon Test
Type I Error Rate
Log Normal Distribution
t-test
Normal Distribution
Bootstrap
Simulation Study
sampling
methodology
testing
Standards
normal distribution

Keywords

  • Bootstrap
  • Cost data
  • Likelihood
  • Log-normal

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Public Health, Environmental and Occupational Health
  • Agricultural and Biological Sciences (miscellaneous)
  • Applied Mathematics
  • Statistics and Probability

Cite this

Methods for comparing the means of two independent log-normal samples. / Zhou, Xiao Hua; Gao, Sujuan; Hui, Siu.

In: Biometrics, Vol. 53, No. 3, 09.1997, p. 1129-1135.

Research output: Contribution to journalArticle

@article{857527fc95354b629a6e803d2da1a014,
title = "Methods for comparing the means of two independent log-normal samples",
abstract = "Standard methods of using the t-test and the Wilcoxon test have deficiencies for comparing the means of two skewed log-normal samples. In this paper, we propose two new methods to overcome these deficiencies: (1) a likelihood-based approach and (2) a bootstrap-based approach. Our simulation study shows that the likelihood-based approach is the best in terms of the type I error rate and power when data follow a log-normal distribution.",
keywords = "Bootstrap, Cost data, Likelihood, Log-normal",
author = "Zhou, {Xiao Hua} and Sujuan Gao and Siu Hui",
year = "1997",
month = "9",
doi = "10.2307/2533570",
language = "English",
volume = "53",
pages = "1129--1135",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Methods for comparing the means of two independent log-normal samples

AU - Zhou, Xiao Hua

AU - Gao, Sujuan

AU - Hui, Siu

PY - 1997/9

Y1 - 1997/9

N2 - Standard methods of using the t-test and the Wilcoxon test have deficiencies for comparing the means of two skewed log-normal samples. In this paper, we propose two new methods to overcome these deficiencies: (1) a likelihood-based approach and (2) a bootstrap-based approach. Our simulation study shows that the likelihood-based approach is the best in terms of the type I error rate and power when data follow a log-normal distribution.

AB - Standard methods of using the t-test and the Wilcoxon test have deficiencies for comparing the means of two skewed log-normal samples. In this paper, we propose two new methods to overcome these deficiencies: (1) a likelihood-based approach and (2) a bootstrap-based approach. Our simulation study shows that the likelihood-based approach is the best in terms of the type I error rate and power when data follow a log-normal distribution.

KW - Bootstrap

KW - Cost data

KW - Likelihood

KW - Log-normal

UR - http://www.scopus.com/inward/record.url?scp=0030767725&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030767725&partnerID=8YFLogxK

U2 - 10.2307/2533570

DO - 10.2307/2533570

M3 - Article

C2 - 9290231

AN - SCOPUS:0030767725

VL - 53

SP - 1129

EP - 1135

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 3

ER -