Using repeated measures to correct correlated measurement errors through orthogonal decomposition. Issue 23 (2nd December 2017)

• Record Type: Journal Article
• Authors: Yu, Chang; Zhang, Sanguo; Friedenreich, Christine; Matthews, Charles E.
• Is Part Of: Communications in statistics. Volume 46:Issue 23(2017)
• Page Start: 11604
• Page End: 11611
• Publication Date: 2017-12-02
• Subjects: Adjusted regression calibration -- correlated measurement errors -- orthogonal decomposition -- physical activity.; Primary: 62F99 -- Secondary: 62P10; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2016.1275693
• Languages: English
• ISSNs: 0361-0926

Abstract

ABSTRACT: In a physical activity (PA) study, the 7-day PA log viewed as an alloyed gold standard was used to correct the measurement error in the physical activity questionnaire. Due to correlations between the errors in the two measurements, the usual regression calibration (RC) may result in a biased estimate of the calibration factor. We propose a method for removing the correlation through orthogonal decomposition of the errors, and then the usual RC can be applied. Simulation studies show that our method can effectively correct the bias.