V-set regression coefficient, 90 v-set regression coefficient – BUCHI NIRCal User Manual
Page 204
NIRCal 5.5 Software Manual
204
NIRCal 5.5 Manual, Version A
3.18.90
V-Set Regression Coefficient
Description
Validation Set Regression Coefficient of Original Property and Predicted
Property (also known as correlation coefficient or Pearson's correlation
coefficient)
Use
To compare with the C-Set regression Coefficient
Method
PCR / PLS
Matrices ID
27
Tip
Should be as close to 1 as possible.
Details
Visible on the calibration curve and in the calibration protocol.
Formula
Related Topic
C-Set Regression Coefficient
,
Original Property
,
Predicted Property
The regression coefficient "r" shows, how well the predicted values match with the reference values
(original property values) on average.
The correlation is rated as acceptable when r > 0.9 is achieved (the error of the conventional reference
method goes into the NIR-calibration via the reference values).
The regression curve can be described better with the slope (a) and intercept (b): f(x)= ax + b