Scholion.ai

Estimating Subgroup Means with Misclassification

Tak K. Mak

Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 1988 · Volume 50 · Issue 1

DOI: 10.1111/j.2517-6161.1988.tb01713.x

Abstract

The present paper considers the estimation of subgroup means in the presence of misclassification errors in allocating individuals to the subgroups. A simple non-iterative estimator resulting from modifying Selen's (1986) estimator is studied when the probabilities of misclassification are independently estimated. In contrast to Selen (1986), an explicit expression for the covariance matrix of the suggested estimator is obtained using special matrix products and matrix differentiation theories. The result is useful in developing large sample theories for hypothesis testing and interval estimation. For the case of double sampling, the problem of deriving a more efficient estimator is considered and a method for allocating sample sizes suggested.

Keywords

categorical data, classification errors, covariance matrix of estimator, subgroup means

Enable JavaScript for citation tools and additional paper interactions.