Determine a model that fits observed data as closely as possible

Often overlooked, error in the specification of a measurement function (often referred to as a "model") can have a significant effect on the results of a measurement method. Modeling, the key step in calibration, is not limited to the point estimation of the parameters of the chosen measurement function (model coefficients). It must also assess the uncertainties surrounding the parameters of this function, in order to estimate the share of uncertainty due to it in the expression of a measurement result. This uncertainty is commonly referred to as "modeling uncertainty". This data sheet covers the mechanisms at work when performing a linear regression, and gives you food for thought when choosing your model.