Chemometrics is the chemical discipline that uses mathematical and statistical
methods to design or select optimal measurement procedures and experiments,
and to provide maximum chemical information by analyzing chemical data.
In modern analytical chemistry and biochemistry, chemometric approaches
have become famous in quantitative and qualitative analysis of samples
from spectroscopic data. The process of data evaluation is called calibration
and will be described in more detail in the following.
In calibration, indirect measurements
are made from samples where the property or the amount of a property to
be evaluated has been pre-determined, usually by an independent technique
or reference measurement. These measurements, along with the pre-determined
property or property levels, comprise a group known as the calibration
set. This set is used to develop a model that relates the property or
property level of a sample to the instrumental measurements. In some cases,
the construction of the model is simple due to a certain relationship,
such as Lambert Beer's Law in the application of UV, IR and NIR spectroscopy.
Unlike spectroscopy, other cases can be much more complex, and it is in
these cases where construction of the model is the time-consuming step.
Once the model is constructed, it can predict sample properties or property
levels based on measurements of new or even unknown samples.
The software focuses on two classical calibration approaches, which
have been widely accepted in the world of analytical chemistry, the quantitative
and the qualitative calibration.
Multivariate calibration allows for the analysis of several measurements
from several samples. This compares to univariate calibration, which involves
the use of a single instrumental measurement to determine a single sample
property. Either method may contribute to a multi-step procedure where
data is calibrated, validated (optional) and further samples predicted
based on the calibration model.
Calibrations are constructed using a wizard which guides the user through
the steps of a calibration. The Calibration Model Wizard supports the user
in creating Univariate Calibrations and Multivariate Calibrations.
After creation of calibration models they will be applied in routine
analysis to predict unknown samples and their concentrations. The software
provides several options to perform predictions and present prediction
results:
The robustness and reliability of the calibration model is reflected by the results of the analysis of variance. Several statistical values are available as shown in the following.
For calibration spectra the following statistics are calculated:
Standard Error of Calibration (SEC)
Root mean square error of Calibration (RMSEC)
Standard Error of cross validation (SECV)
Root mean square error of cross validation (RMSECV)
For validation spectra the following statistics are calculated:
Standard error of prediction (SEP)
Root mean square error of prediction (RMSEP)
References
K. Danzer and L.A. Currie
Guidelines for calibration in analytical chemistry
Part 1. Fundamentals and single component calibration
Pure & Applied Chemistry, 70 (1998) 993-1014.
K. Danzer, M. Otto and L.A. Currie
Guidelines for calibration in analytical chemistry
Part 2. Multispecies calibration
Pure & Applied Chemistry, 76 (2004) 1215-1225.
M. Otto
Chemometrics, Wiley-VCH, 1999
H. Martens, T. Naes
Multivariate Calibrtion, Wiley, 1989
