Variance Estimation and Precision Results - Residuals Tab

Select the Residuals tab of the Variance Estimation and Precision Results dialog box to access the options described here.

Residual type

Select the Residual type that you want to use in the plots. You can use either Studentized residuals or the Raw residuals.

Residuals

Click the Residuals button to produce a spreadsheet of the observed, predicted, and residual values; the type of residual that will be used can be chosen in the Residual type group box (see above).

Normal probability plot

Click the Normal probability plot button to produce a normal probability plot of the residual values; the type of residual that will be used can be chosen in the Residual type group box (see above).

Residuals vs. predicted.

Click the Residuals vs. predicted button to produce a 2D scatterplot of the residual versus predicted values; the type of residual that will be used can be chosen in the Residual type group box (see above).

Residuals in sequence order

Click the Residuals in sequence order button to produce a 2D scatterplot of the residual values versus the case numbers (of the observations in the plot). This plot is useful in order to detect any serial correlation between the residuals for consecutive observations.

Residuals vs selected variable

Click the Residuals vs selected variable to display a single variable selection dialog box and select a variable to plot against the residual values in a 2D scatterplot. You can use this option to create a scatterplot of the residuals vs. the observed values (by selecting the response variable) or the residuals vs. a specific factor.

Observed by predicted

Click the Observed by predicted button to produce a 2D scatterplot of the predicted values versus the actual (observed) values.

Residuals vs selected effect

Click the Residuals vs selected effect button to display a single variable selection dialog box and select an effect (either fixed or random) to plot against the residual values in a 2D scatterplot. This plot enables you to visually inspect the assumption of homogeneity of variability within effects.