Metrohm Vision – Tutorial - Manual (Page 25)

 5 Quantitative Model Development
18
 NIRS Vision Tutorial – Operation
5 Quantitative Model Development
Quantitative Analysis mode is the part of Vision where calibration equations
are developed and tested. The created calibration equations can be used in
Routine Analysis to predict unknown sample composition in real time.
Vision offers two types of calibration development: Multilinear Regression
(MLR) and Partial Least Squares (PLS).
1.6 Multilinear Regression
Multilinear Regression (MLR) is a least squares method that uses spectral
information at one or several wavelengths. In the simplest case (one
wavelength), the method reduces to simple Linear Regression.
1.6.1 Sample Selection
When developing a calibration equation to predict a constituent’s
concentration in a sample, Sample Selection must be performed on the
spectral dataset for each equation to be developed. Equations/models must be
developed for 1 constituent at a time and for each constituent you must go
through Sample Selection
Select a Product
1. Click on Mode  Quantitative Analysis  Sample Selection.
2. Double click on the “xdsstart” project
3. Click and drag the product “MCC” onto the red EQ and release the
mouse button.
4. In the Select Appropriate Constituent dialog constituent Caffeine is
listed. Enter:
 Unit %
5. Click on [OK]
6. In the Calibration Equation dialog box, add a proper equation name
(e. g. caffeine_mlr), click on [OK]
Notice Caffeine now appears under the red Eq and the cuvette is gray.