Data matrix used for modelling

Scaling converts the original descriptor matrix to a normalized data matrix, which hereafter will be denoted by The columns in X are the scaled descriptor variables, see Fig. 15.9. 

The scaled data matrix is usually denoted by X in Factor analysis and Principal components analysis. Unfortunately, the same symbol is used to denote the model matrix in response surface (linear regression) modelling. Hopefully, the reader will excuse this overlap in nomenclature. 

3 J Mathematical derivation of the principal components model. Loading vectors, score vectors 

Geometric illustrations are used to show the mathematical aspects of the modelling process. It is shown that these models are related to eigenvalues and eigenvectors. See Fig 15.10 for the definitions. 

Let M be the point in the descriptor space which corresponds to the average value of each descriptor  

---