Estimating precision, repeatability, and reproducibility from Gaussian and non- Gaussian data: a mixed models approach

Assam Pryseley, Koen Mintiens, Katia Knapen, Yves Van der Stede and Geert Molenberghs

Journal of Applied Statistics, 2010, vol. 37, issue 10, 1729-1747

Abstract: Quality control relies heavily on the use of formal assessment metrics. In this paper, for the context of veterinary epidemiology, we review the main proposals, precision, repeatability, reproducibility, and intermediate precision, in agreement with ISO (international Organization for Standardization) practice, generalize these by placing them within the linear mixed model framework, which we then extend to the generalized linear mixed model setting, so that both Gaussian as well as non-Gaussian data can be employed. Similarities and differences are discussed between the classical ANOVA (analysis of variance) approach and the proposed mixed model settings, on the one hand, and between the Gaussian and non-Gaussian cases, on the other hand. The new proposals are applied to five studies in three diseases: Aujeszky's disease, enzootic bovine leucosis (EBL) and bovine brucellosis. The mixed-models proposals are also discussed in the light of their computational requirements.

Keywords: accuracy; analysis of variance; Aujeszky's disease; bias; bovine brucellosis; enzootic bovine leucosis; generalized linear mixed models; linear mixed models; quality control (search for similar items in EconPapers)
Date: 2010
References: View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.tandfonline.com/doi/abs/10.1080/02664760903150706 (text/html)
Access to full text is restricted to subscribers.

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:taf:japsta:v:37:y:2010:i:10:p:1729-1747

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/CJAS20

DOI: 10.1080/02664760903150706

Access Statistics for this article

Journal of Applied Statistics is currently edited by Robert Aykroyd

More articles in Journal of Applied Statistics from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().