Simple view on Simple Interval Calculation (SIC) method

Various regression methods and new interval approach are two different manners for solution of multivariate calibration problems. Both of these techniques aim to construct a model that associates multiple predictor variables with response and to predict the unknown response value for new predictors. When applied to the same data set both methods provide a researcher with reach information and results that supplement each other, despite the fact that different assumptions regarding the nature of error stand behind the two approaches and different mathematical techniques are applied. The goal of this work is to draw the attention of analysts to the outcomes that may be obtained under the assumption of error finiteness and to compare the results, provided by the interval approach with those yielded by regression modeling.