Application of Bayesian model averaging using a fixed effects model with linear drift for the analysis of key comparison CCM.P-K12
- 104 Downloads
In this paper, we extend a recently proposed procedure for the analysis of key comparison data to the case where the traveling standard shows a linear drift. The procedure is based on a fixed effects model and assumes that a certain number of the laboratories measure without bias. The analysis is carried out by applying Bayesian model averaging to all possible different models, each assuming zero biases for a different subset of laboratories. The method is applied to the key comparison CCM.P-K12.
Keywordsfixed effects model Bayesian averaging method of (weighted) least squares bias
- 5.Mutual Recognition of National Measurement Standards and of Calibration and Measurement Certificates Issued by National Metrology Institutes (MRA), BIPM (1999).Google Scholar
- 6.Evaluation of Measurement Data. Supplement 1 to the Guide to the Expression of Uncertainty in Measurement. Propagation of Distributions Using a Monte Carlo Method, Joint Committee for Guides in Metrology, JCGM 101:2008 BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP, and OIML.Google Scholar
- 11.D. R. White, “On the analysis of measurement comparisons,” ibid., 122–131.Google Scholar