Random Regression Forest

Select menu Stats | Multivariate Analysis | Random Regression Forest

Use this to form a random classification forest. This uses the BRFOREST procedure to randomly select variates and a subset of units of the training data set, and creates optimal Regression Trees for each selection. The prediction is then performed using the consensus of the predictions from the individual regression trees. See the help on the Regression Trees menu for more information on how the trees are produced.

Available data

This lists data structures appropriate to the current input field. The contents will change as you move from one field to the next. Double-click a name to copy it to the current input field; alternatively, you can type the name directly into the input field.

Y-variate

Specify a response variate for the regression.

X-variables

Specifies the independent (x) variables available for constructing the tree. The variables can be factors or variates. You can transfer multiple selections from Available data by holding the Ctrl key on your keyboard while selecting items, then click to move them all across in one action.

Save forest in:

Specifies an identifier name to save the resulting forest in. The forest will be saved within a Genstat pointer to the individuals trees.

Save

The button opens the Save dialog which allows you to store the store the out of bag predictions, out of bag error, and importance of the variables from the forest.

Further output

The button opens the Further Output dialog which allows you to display information on the forest from the analysis.

Predict

The button opens the Predict dialog which allows you to store or display the predictions for the current data set or for a new set of observations.