TMPLS Results - Advanced Tab

Select the Advanced tab of the TMPLS Results dialog box to access the options described here.

Prediction

Following are descriptions of the options available in this group box.

Element Name	Description
Use original scale	Select this check box to generate predictions and residuals on the scale of the original or raw data. If not selected, predictions and residuals will be based on the normalized/scaled data.
X Predictions	Click this button to produce a spreadsheet of the X predictions for the process variables.
Y Predictions	Click this button to produce a spreadsheet of the Y predictions for the process variables.
X Residuals	Click this button to produce a spreadsheet of residuals for the process variables. Residuals are defined as the difference between the original variables and the predictions of the TMPLS model. In other words, residuals are the unmodeled parts of the data that could not be matched by the predictions of the model. Large residuals are indications of abnormality in the data that cannot be sufficiently predicted by the model. The ability to detect outliers is a useful feature of TMPLS that can be utilized for process monitoring and quality control.
Y Residuals	Click this button to produce a spreadsheet of residuals for the dependent (quality) variables. Residuals are defined as the difference between the original variables and the predictions of the TMPLS model. In other words, residuals are the unmodeled parts of the data that could not be matched by the predictions of the model. Large residuals are indications of abnormality in the data that cannot be sufficiently predicted by the model. The ability to detect outliers is a useful feature of TMPLS that can be utilized for process monitoring and quality control.
Coefficients	Click this button to generate a spreadsheet of the coefficients for the TMPLS model (see PCA and PLS Technical Notes). The coefficients of a TMPLS model relate the u-scores for the dependent (quality) variables to the t-scores of the independent (continuous and/or categorical process variables) variables.
Eigenvalues	Click this button to generate a spreadsheet of eigenvalues for the PC model.
X Weights (w)	Click this button to produce a spreadsheet containing the weights of the independent (process) variables.
Scree plot	Click this button to create an eigenvalue scree plot  (Cattell, 1966) for the extracted principal components. By default only the extracted eigenvalues are included but you can extend this number (up to the maximum number of the eigenvalues) using Number of eigenvalues option below.
Number of eigenvalues	In this box, specify how many eigenvalues to be included in the scree plot.
X Scores (t)	Click this button to produce a spreadsheet of t-scores. The t-scores are the representation of the independent (process) variables in the new coordinate system, i.e., the system of the principal components.
X Loadings (p)	Click this button to create the matrix of the p-loading factors for the principal components in spreadsheet format. The loading factors determine the orientation of the principal component axes with respect to the original coordinate system (defined by the process variables). Loading factors are used to analyze the influence of the original variables in determining the TMPLS model.
X Descriptives	Click this button to produce a spreadsheet of various statistics of the original independent (process) variables such as number of valid cases, means, standard deviations, and scale.
X Scaled data	Click this button to generate a spreadsheet of the pre-processed (scaled) independent (process) variables. Pre-processing involves the application of a linear transformation which transforms the original data set to a new set of variables with zero mean and unit (or user specified) standard deviation.
D-To-Model (X)	Click this button to produce a spreadsheet of distance-to-model for all batches in the data set. Distance-to-model plays an important role in process control since it measures the squared perpendicular distance of an observation from the normal plane. Distance-to-model is used as an indication of whether a batch is within the domain of normality. Hence, it can be used for detecting outliers.
D-To-Model (X)	Click this button to produce the D-To-Model in line plot format.
D-To-Model (X)	Click this button to produce the D-To-Model in histogram format.
Variable	Use the options in this group box to generate data for variable trajectories in the form of plots and spreadsheet format.
Trajectory	Click this button to generate a line plot of process variable trajectories against time for all batches in the data set.
(Trajectory Spreadsheet)	Click this button to generate the same data in spreadsheet format.
Trajectory variable list	Use this drop-down menu to select a variable for which you want to print a trajectory plot or spreadsheet.
Limits as regions	Select this checkbox to display the area inside the control and warning limits of the variable trajectory chart as different colors in order to more easily identify the in-control and out-of-control batches.
Control	In this box, specify the control limits. This value is multiplied by the computed standard deviation for defining the upper and lower limits.
Warning	Select this check box to include an upper and a lower warning limit in the variable trajectory chart. Specify the warning limit in the adjacent box. This value will be multiplied by the computed standard deviation to define the lower and upper warning limits.
Y Scores (u)	Click this button to generate a spreadsheet of u-scores of the dependent (quality) variables. The u-scores are simply the representation of the original dependent variables in the new coordinate system.
Y Loadings (p)	Click this button to generate a spreadsheet of the p-loadings for the dependent (quality) variables. The p-loadings determine the orientation of the principal components with respect to the original coordinate system. They are often used to determine how various process variables influence the TMPLS system and, hence, the batch process itself.
Y Descriptives (g)	Click this button to display a spreadsheet of various statistics of the original dependent (quality) variables such as number of valid cases, means, standard deviations, and scale.
Y Scaled data	Click this button to generate a spreadsheet of the preprocessed (scaled) dependent (quality) variables. Preprocessing involves the application of a linear transformation which transforms the original data set to a new set of variables with zero mean and unit (or user specified) standard deviation.
D-To-Model (Y)	Click this button to display a spreadsheet of distance-to-model for all batches in the data set. Distance-to-model plays an important role in process control since it measures the squared perpendicular distance of an observation from the normal plane. Distance-to-model is used as an indication of whether a batch is within the domain of normality. Hence, it can be used for detecting outliers.
D-To-Model (Y)	Click this button to display the D-To-Model in line plot format.
D-To-Model (Y)	Click this button to display the D-To-Model in histogram format.