The effect of errors in diagnosis and measurement on the estimation of the probability of an event

Joel E. Michalek; Ram C. Tripathi

doi:10.1080/01621459.1980.10477539

The effect of errors in diagnosis and measurement on the estimation of the probability of an event

Joel E. Michalek, Ram C. Tripathi

Epidemiology & Biostatistics

Research output: Contribution to journal › Article › peer-review

17 Scopus citations

Abstract

This article investigates the effect of misclassification and measurement error in the basic data on the asymptotic bias and efficiency of the logistic regression (LR) and normal discrimination (ND) classification procedures. The effect of misclassification in a single binary independent variable on the bias and efficiency of both procedures is also presented. Typically, asymptotic bias increases and efficiency decreases as misclassification and measurement error increase. The performance of LR relative to ND is shown to be better in the presence of error than without error.

Original language	English (US)
Pages (from-to)	713-721
Number of pages	9
Journal	Journal of the American Statistical Association
Volume	75
Issue number	371
DOIs	https://doi.org/10.1080/01621459.1980.10477539
State	Published - Sep 1980

Keywords

Asymptotic relative efficiency
Logistic regression
Normal discrimination

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1080/01621459.1980.10477539

Fingerprint Dive into the research topics of 'The effect of errors in diagnosis and measurement on the estimation of the probability of an event'. Together they form a unique fingerprint.

Cite this

@article{87e0a520315e46e6b6b3618371751856,

title = "The effect of errors in diagnosis and measurement on the estimation of the probability of an event",

abstract = "This article investigates the effect of misclassification and measurement error in the basic data on the asymptotic bias and efficiency of the logistic regression (LR) and normal discrimination (ND) classification procedures. The effect of misclassification in a single binary independent variable on the bias and efficiency of both procedures is also presented. Typically, asymptotic bias increases and efficiency decreases as misclassification and measurement error increase. The performance of LR relative to ND is shown to be better in the presence of error than without error.",

keywords = "Asymptotic relative efficiency, Logistic regression, Normal discrimination",

author = "Michalek, {Joel E.} and Tripathi, {Ram C.}",

year = "1980",

month = sep,

doi = "10.1080/01621459.1980.10477539",

language = "English (US)",

volume = "75",

pages = "713--721",

journal = "Journal of the American Statistical Association",

issn = "0162-1459",

publisher = "Taylor and Francis Ltd.",

number = "371",

}

TY - JOUR

T1 - The effect of errors in diagnosis and measurement on the estimation of the probability of an event

AU - Michalek, Joel E.

AU - Tripathi, Ram C.

PY - 1980/9

Y1 - 1980/9

N2 - This article investigates the effect of misclassification and measurement error in the basic data on the asymptotic bias and efficiency of the logistic regression (LR) and normal discrimination (ND) classification procedures. The effect of misclassification in a single binary independent variable on the bias and efficiency of both procedures is also presented. Typically, asymptotic bias increases and efficiency decreases as misclassification and measurement error increase. The performance of LR relative to ND is shown to be better in the presence of error than without error.

AB - This article investigates the effect of misclassification and measurement error in the basic data on the asymptotic bias and efficiency of the logistic regression (LR) and normal discrimination (ND) classification procedures. The effect of misclassification in a single binary independent variable on the bias and efficiency of both procedures is also presented. Typically, asymptotic bias increases and efficiency decreases as misclassification and measurement error increase. The performance of LR relative to ND is shown to be better in the presence of error than without error.

KW - Asymptotic relative efficiency

KW - Logistic regression

KW - Normal discrimination

UR - http://www.scopus.com/inward/record.url?scp=5844300663&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=5844300663&partnerID=8YFLogxK

U2 - 10.1080/01621459.1980.10477539

DO - 10.1080/01621459.1980.10477539

M3 - Article

AN - SCOPUS:5844300663

VL - 75

SP - 713

EP - 721

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 371

ER -