Minimum variance estimators for misclassification probabilities in discriminant analysis

James D. Broffitt and J. S. Williams

Journal of Multivariate Analysis, 1973, vol. 3, issue 3, 311-327

Abstract: Let [alpha](n1, n2) be the probability of classifying an observation from population [Pi]1 into population [Pi]2 using Fisher's linear discriminant function based on samples of size n1 and n2. A standard estimator of [alpha], denoted by T1, is the proportion of observations in the first sample misclassified by the discriminant function. A modification of T1, denoted by T2, is obtained by eliminating the observation being classified from the calculation of the discriminant function. The UMVU estimators, and , of ET1 = [tau]1(n1, n2) and ET2 = [tau]2(n1, n2) = [alpha](n1 - 1, n2) are derived for the case when the populations have multivariate normal distributions with common dispersion matrix. It is shown that and are nonincreasing functions of D2, the Mahalanobis sample distance. This result is used to derive the sampling distributions and moments of and . It is also shown that [alpha] is a decreasing function of [Delta]2 = ([mu]1 - [mu]2)'[Sigma]-1([mu]1 - [mu]2). Hence, by truncating and (or any estimator) at the value of [alpha] for [Sigma] = 0, new estimators are obtained which, for all samples, are as close or closer to [alpha].

Keywords: Estimating; misclassification; probabilities; classification; linear; discriminant; function; minimum; variance; estimation (search for similar items in EconPapers)
Date: 1973
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/0047-259X(73)90044-4
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:jmvana:v:3:y:1973:i:3:p:311-327

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/supportfaq.cws_home/regional
https://shop.elsevie ... _01_ooc_1&version=01

Access Statistics for this article

Journal of Multivariate Analysis is currently edited by de Leeuw, J.

More articles in Journal of Multivariate Analysis from Elsevier
Bibliographic data for series maintained by Catherine Liu ().