Boosted Trees Specifications

Click the OK button in the Boosted Trees Startup Panel to display the Boosted Trees Specifications dialog box. Use these options to specify the parameters for the analysis. The specific options available on the Quick and Advanced tabs will depend on whether the current analysis is a Regression Analysis or Classification Analysis, as specified on the Startup Panel Quick tab. For classification problems, a Classification tab is also available.

OK

Click the OK button to begin the analysis; the Computing... dialog box will be displayed where you can observe the iterations, i.e., as each successive tree is computed. See also the Overview or Computational Details for details.

Partitioning of data on OK

If no Test sample variable and code is specified on the Advanced tab, STATISTICA will create by default a random sub-sample of 30% of the observations (cases) in the data and treat them as a test sample in order to evaluate the fit of the model over successive iterations in this separate (hold-out) sample. You can adjust this value via the Random test data proportion option on the Advanced tab of this dialog box. Given the default proportion (30%) of test cases, this leaves the remaining 70% of the observations for the analyses via stochastic gradient boosting (e.g., for the selection of samples for consecutive boosting steps). By default, the program will choose the specific solution (with the specific number of simple boosted trees) that yields the absolute smallest error (misclassification rate) over all boosting iterations. Use the Test sample options on the Advanced tab if you want to select a specific sub-sample for the test (hold-out) sample.

Cancel

Click the Cancel button to close the dialog box without performing an analysis, and to display the Boosted Trees Startup Panel.

Options

Click the Options button to display the Options menu.