Bayesian uncertainty analysis for a regression model versus application of GUM Supplement 1 to the least-squares estimate

Application of least-squares as, for instance, in curve fitting is an important tool of data analysis in metrology. It is tempting to employ the supplement 1 to the GUM (GUM-S1) to evaluate the uncertainty associated with the resulting parameter estimates, although doing so is beyond the specified scope of GUM-S1. We compare the result of such a procedure with a Bayesian uncertainty analysis of the corresponding regression model. It is shown that under certain assumptions both analyses yield the same results but this is not true in general. Some simple examples are given which illustrate the similarities and differences between the two approaches.

[1]  C. Elster.  Calculation of uncertainty in the presence of prior knowledge , 2007 .

[2]  Clemens Elster,et al.  Draft GUM Supplement 1 and Bayesian analysis , 2007 .

[3]  Clemens Elster,et al.  Bayesian uncertainty analysis under prior ignorance of the measurand versus analysis using the Supplement 1 to the Guide: a comparison , 2009 .

[4]  Christian Genest,et al.  Combining Probability Distributions: A Critique and an Annotated Bibliography , 1986 .

[5]  G. Kyriazis,et al.  Comparison of GUM Supplement 1 and Bayesian analysis using a simple linear calibration model , 2008 .

[6]  Alistair B. Forbes.  NONLINEAR LEAST SQUARES AND BAYESIAN INFERENCE , 2009 .

[7]  R Willink.  Estimation and uncertainty in fitting straight lines to data: different techniques , 2008 .

[8]  George E. P. Box,et al.  Bayesian Inference in Statistical Analysis: Box/Bayesian , 1992 .

[9]  Peter M. Harris,et al.  MEASUREMENT UNCERTAINTY EVALUATION ASSOCIATED WITH CALIBRATION FUNCTIONS , 2009 .

[10]  J. G. Saw,et al.  The Distribution of Linear Combinations of t-Variables , 1978 .

[11]  Ignacio Lira,et al.  Equivalence of alternative Bayesian procedures for evaluating measurement uncertainty , 2010 .

[12]  Raghu N. Kacker,et al.  On use of Bayesian statistics to make the Guide to the Expression of Uncertainty in Measurement consistent , 2003 .

[13]  R Willink,et al.  Forming a comparison reference value from different distributions of belief , 2006 .

[14]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[15]  L. Wasserman,et al.  The Selection of Prior Distributions by Formal Rules , 1996 .

[16]  C. Elster,et al.  Probabilistic and least-squares inference of the parameters of a straight-line model , 2007 .

[17]  Blaza Toman,et al.  Assessment of measurement uncertainty via observation equations , 2007 .