PLS Results - Observational Tab

Partial Least Squares (PLS)

Select the Observational tab of the PLS Results dialog box to access options to display and plot various observational statistics, such as predicted and residual values, factor scores, etc. For additional details about these options, see the Introductory Overview or the Partial Least Squares (PLS) Index.

Sample. Select the an option button under Sample to specify which type of sample to base the predicted and residual values.

 Analysis/Cross-val/Both/Prediction. You can display and plot predicted and residual values for all observations that were used to compute the current results (select Analysis), all observations that were not used to compute the current results, but have valid data for all predictors and the dependent variable (select Cross-val(idation)), both the Analysis and Cross-validation samples (select Both), or display and plot predicted values for all cases that have valid data for the predictor variables, but missing data for at least one dependent variable (select Prediction). If these options are not available, no valid cases in the prediction sample were found, and/or no cross-validation sample was specified via the PLS Quick Specs - Options tab, or via the SAMPLE keyword in the PLS Analysis Syntax Editor.

Residuals. Click the Residuals button to display a spreadsheet with the raw residuals, computed for the respective number of components as specified in the Number of components field on the PLS Results dialog box.

Predicted values. Click the Predicted values button to display a spreadsheet with the predicted values for each dependent (response) variable, computed for the specified Number of components (see PLS Results).

Scores. Click the Scores button to display a spreadsheet with the factor scores for each dependent (response) variable, computed for the specified Number of components (see PLS Results).

Raw data. Click the Raw data button to display a spreadsheet with the raw data values for the predictor (X) variables as well as the dependent (response) variables. This option is particularly useful for reviewing the coded values for categorical predictor variables in the design matrix (see also the GLM Introductory Overview topic Sigma-Restricted and Overparameterized Model for details).

Histograms. Use the options under Histograms to create histograms of the values for any of the dependent (response) variables in the analysis, or for any of the predicted or residual values for the dependent (response) variables.

Variable. Select the variable/value to be plotted in the histogram in the Variable drop-down list.

Plot of selected item. Click the Plot of selected item button to display a histogram of the variable/value you selected (see Variable above).

Bin #. Enter the number of categories in the histogram (the number of bins) in the Bin # field.

Normal probability plots. Use the options under Normal probability plots to create normal probability plots of the values for any of the dependent (response) variables in the analysis, or for any of the predicted or residual values for the dependent (response) variables.

Variable. Select the variable/value to be plotted in the normal probability plot in the Variable drop-down list.

Plot of selected item. Click the Plot of selected item button to display a normal probability plot of the variable/value you selected (see Variable above).

Observational scatterplots. Use the options under Observational scatterplots to create 2D scatterplots of the values for any of the dependent (response) variables in the analysis, or for any of the predicted or residual values for the dependent (response) variables.

 Y. Select the variable/value to be plotted on the y-axis of the scatterplot in the Y drop-down list.

X. Select the variable/value to be plotted on the x-axis of the scatterplot in the X drop-down list.

Plot of selected items. Click the Plot of selected item button to display a scatterplot of the variables/values you selected (see Y, X above)

For additional details about the computations in PLS, see Computational Approach or the Partial Least Squares (PLS) Index.