Error-free milestones in error prone measurements

DOI10.1214/08-AOAS233zbMATH Open1196.62048arXiv0910.1233OpenAlexW2073170980WikidataQ35857697 ScholiaQ35857697MaRDI QIDQ985011FDOQ985011

Authors: Dylan S. Small, Paul R. Rosenbaum 

Publication date: 20 July 2010

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: A predictor variable or dose that is measured with substantial error may possess an error-free milestone, such that it is known with negligible error whether the value of the variable is to the left or right of the milestone. Such a milestone provides a basis for estimating a linear relationship between the true but unknown value of the error-free predictor and an outcome, because the milestone creates a strong and valid instrumental variable. The inferences are nonparametric and robust, and in the simplest cases, they are exact and distribution free. We also consider multiple milestones for a single predictor and milestones for several predictors whose partial slopes are estimated simultaneously. Examples are drawn from the Wisconsin Longitudinal Study, in which a BA degree acts as a milestone for sixteen years of education, and the binary indicator of military service acts as a milestone for years of service.

Full work available at URL: https://arxiv.org/abs/0910.1233

Recommendations

• Measurement errors in the binary instrumental variable model
• QUANTILE REGRESSION WITH MISMEASURED COVARIATES
• Quantile regression with measurement error
• Nonlinear and Nonparametric Regression and Instrumental Variables
• Instrumental Variable Treatment of Nonclassical Measurement Error Models

zbMATH Keywords

instrumental variablesattenuationfull matchingquestionnaire designerrors in measurementnonbipartite matching

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Applications of statistics to social sciences (62P25) Nonparametric inference (62G99)

Cites Work

• Estimates of Location Based on Rank Tests
• Robust Statistics
• Title not available (Why is that?)
• Full Matching in an Observational Study of Coaching for the SAT
• Title not available (Why is that?)
• Title not available (Why is that?)
• Robust, Accurate Confidence Intervals with a Weak Instrument: Quarter of Birth and Education
• Title not available (Why is that?)
• Computing Minimum-Weight Perfect Matchings
• Title not available (Why is that?)
• Title not available (Why is that?)
• The Fitting of Straight Lines When both Variables are Subject to Error
• Efficient algorithms for finding maximum matching in graphs
• The Fitting of Straight Lines if Both Variables are Subject to Error
• Title not available (Why is that?)
• Rank tests for bivariate symmetry
• Permutation tests for reflected symmetry
• On Certain Methods of Estimating the Linear Structural Relation
• Exact, Nonparametric Inference When Doses Are Measured With Random Errors
• Differential effects and generic biases in observational studies
• A Characterization of Population Weighted-Symmetry and Related Results
• On the Theory of Rank Order Tests for Location in the Multivariate One Sample Problem
• On a Further Robustness Property of the Test and Estimator Based on Wilcoxon's Signed Rank Statistic

Cited In (4)

• Better stability with measurement errors
• Reliable measurement and influence of measurement errors on the balance algorithm of physical parameter (application on a rectification vacuum column)
• Relaxation penalties and priors for plausible modeling of nonidentified bias sources
• Markers for error-corrupted observations

Uses Software

• Blossom IV
• optmatch

This page was built for publication: Error-free milestones in error prone measurements

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q985011)