PLOT

The plots available for the selected model are predefined and will be highlighted in the menu function after the model has been calculated.

 

Bi-plot (scores and loadings)

Scores and loadings for one PC against another PC and a bi-plot showing both parameters together.

 

Loadings (line plot)

Loadings for component F (PCA) or mode one to three (PARAFAC/2).

 

Explained variance

Percent variance explained for the fitted and the cross validated model for all components.

 

Influence plot

Leverage vs. residuals. Leverage is a measure of how extreme a data point or a variable is compared to the majority. In PCA and PLS, leverage can be interpreted as the distance between a projected point (or projected variable) and the model centre. Residuals are a measure of the variation that is not taken into account by the model (Xpre-processed or raw data = Xmodelled + Residuals).

 

Modelling plot

Raw data, residuals and model for the third mode variables (K).

 

T vs. U plot

This plot shows the relationship between the projection of your samples in the X-space (T axis) and the projection of your samples in the Y-space (U axis) for F components. The T elements are the scores for samples in the X-matrix whereas U corresponds to the Y-matrix score values.

 

Predicted vs. Measured

Predicted Y-values for the samples plotted against the actual Y-values of the samples. Statistics (i.e. model parameters) – R (correlation), slope (of regression line), intercept (at ordinate axis) and RMSEP (Root Mean Square Error of Prediction) are shown in the plot. RMSEP is a measurement of the average difference between predicted and measured response values, at the prediction or validation stage.