Stat > Predictive
Analytics > CART®
Regression > Graphs
Select the graphs that you want to display for the analysis.
Residuals for
plots
Specify the type of residuals to display on the boxplot of residual plots.
Regular: By default, the boxplot displays regular residuals.
Percent: Specify to display the percentage residuals on the boxplot.
Tree
diagram
The tree diagram shows the optimal tree. You can right-click the diagram to switch between the detailed and node split view. The detailed view of the tree includes the mean, standard deviation, and total count of each node. The node split view shows the variable that splits the data at each node.
Number of terminal nodes plot
R-squared vs. number of terminal nodes
plot
The R-squared vs. number
of terminal nodes plot shows the relationship between R-squared value and tree size when the Least
squared error node splitting method is selected.
Mean
absolute deviation vs. number of terminal nodes plot
The Mean absolute deviation vs. number
of terminal nodes plot shows the relationship between mean absolute deviation value and tree size when the Least absolute deviation node splitting method is selected.
Variable importance chart
The Variable importance
chart shows the relative importance of the predictors. You can choose whether to display all or some of the important variables. Variables are important when they are used as primary and surrogate splitters.
Display all important variables: By default, this chart displays all important variables.
Display a percentage of important
variables: Specify the percentage of important variables to display. Enter a value between 0 and 100.
Display all predictor variables: Display all predictors whether or not they are important variables.
Fitted vs.
actual response value plot
The Fitted vs. actual response value plot shows the fitted Y (response) values versus the actual Y (response) values for both the training and test data sets.
Boxplot of
residuals
The Boxplot of residuals shows the residual values or the percent residuals for both the training and test data sets.
Terminal node order
You can specify the order of the terminal nodes for the following plots: Mean squared error vs. terminal node plot, Mean absolute deviation vs. terminal node plot, Boxplot of response by terminal node, and the Residual vs. terminal node plot.
Ascending mean squared error
The terminal nodes are ordered in increasing mean squared error values, from lowest to highest, when the Least
squared error node splitting method is selected.
Ascending mean absolute deviation
The terminal nodes are ordered in increasing mean squared error values, from lowest to highest, when the Least absolute deviation node splitting method is selected.
No
sorting
The terminal nodes are in the same order as in the tree diagram, such as 1, 2, 3, and so on.
Mean squared error or mean absolute deviations vs number of terminal nodes plot
Mean squared
error vs. terminal node plot
The Mean squared error vs. terminal nodes plot shows the mean squared error for each terminal node when the Least
squared error node splitting method is selected. This chart displays in ascending mean squared error order by default.
Mean
absolute deviation vs. terminal node plot
The Mean absolute deviation vs. terminal nodes plot shows the mean absolute deviation for each terminal node when the Least absolute deviation node splitting method is selected. By default, this chart displays in ascending mean squared error order.
Boxplot
of response by terminal node
The Boxplot of response by terminal node plot shows a boxplot of responses for each terminal node. This chart displays the terminal nodes in ascending mean squared error order or ascending mean absolute deviation order, by default.
Residual vs.
terminal node plot
The Residuals vs. terminal node plot shows a plot of residuals or percent residuals for each terminal node. This chart displays the terminal nodes in ascending mean squared error order or ascending mean absolute deviation order, by default. The top panel is for the training data set, and the bottom panel is for the test data set.
