We compare three approaches for quantifying uncertainty using a measurement equation: the International Organization for Standardization (ISO) Guide to the Expression of Uncertainty in Measurement (GUM), draft GUM Supplement 1, and Bayesian statistics. We use the measurement equation for simple linear calibration as an illustration. It includes both TypeA and TypeB input variables. We consider three scenarios: (i) the measurement equation is linear. (ii) the measurement equation is non-linear and the TypeB input variables have normal distributions, and (iii) the measurement equation is non-linear and the Type B input variables have rectangular distributions. We consider both small and large uncertainties for the Type B input variables. We use each of the three approaches to quantify the uncertainty in measurement for each of the cases considered. Based on this study and the original publications, we discuss the merits and limitations of each approach.
Pub Type: Journals
Bayesian statistics, linear calibration, standard uncertainty, uncertainty intervals, uncertainty in measurement