Uncertainty of Measurement in the Linear Regression Method Part 2. Uncertainty Bands of the Regression Straight-Line for the Correlated Data of Y Variable
This is the continuation of authors’ works on the description of the accuracy of various straight-line cases determined from the results of linear regression measurements. In the first work, the essence, criteria and dependencies of the regression method were examined, as well as simulated examples of determining simple uncertainty bands fitted to measured points with uncorrelated ordinates. The GUM Guide was referred to and the B type uncertainty not discussed yet in the literature about the application of the regression method in measurements was taken into account. This work discusses determining the equation of a simple regression and its uncertainty bands from measuring points with ordinates with autocorrelation. This is illustrated by examples with precisely known abscissa and ordinates with different correlation variants, and absolute and relative uncertainty types A and B. Proposed is the extended method for assessing the accuracy of simple regression takes into account both the correlation of the Y variable data and the impact of type B uncertainty in routine measurements.