Theory & Methods: Jackknifing the Mantel-Haenszel Estimator in Sparse Contingency Tables

This article investigates the performance of two jackknife techniques under an asymptotic model in which the number of 2 × 2 tables increases but the possible marginal configurations remain fixed. These approaches are applied to the Mantel–Haenszel estimator, or transformed versions of this estimato...

Full description

Saved in:
Bibliographic Details
Published in:Australian & New Zealand journal of statistics 1998-12, Vol.40 (4), p.433-441
Main Authors: Pigeot, I., Strugholtz, H.
Format: Article
Language:English
Subjects:
bias reduction
common odds ratio
jackknife
sparse data
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article investigates the performance of two jackknife techniques under an asymptotic model in which the number of 2 × 2 tables increases but the possible marginal configurations remain fixed. These approaches are applied to the Mantel–Haenszel estimator, or transformed versions of this estimator, respectively. The resulting jackknife estimators are shown to be consistent for the common odds ratio. Their asymptotic distributions are derived; they can be used for constructing appropriate sparse‐data confidence intervals.
ISSN:1369-1473
1467-842X
DOI:10.1111/1467-842X.00048