Statistics Decision Trees
The Statistics Decision Trees Module, starting from a dataset, allows you to identify groups, detect relationships and predict future events. It integrates decision tree and classification structures that allow categorical results to be represented intuitively through tree graphs, simplifying the understanding of analysis results.
The Decision Trees Module includes four tree growth algorithms and gives you the opportunity to try out different types of algorithms to find the one that best suits your data. It offers advanced tree structure creation techniques for classification directly in the Statistics for Data Analysis environment.
The four tree growth algorithms include:
CHAID-A multi-tree statistics algorithm that allows you to visualise data quickly and efficiently, creating segments and profiles based on the desired results.
Exhaustive CHAID-A modification of the CHAID algorithm, which examines all possible separations for each predictor.
Classification and regression trees (C&RT)-A full binary tree algorithm that partitions the data and produces precise homogeneous subsets.
QUEST-A statistical algorithm that selects variables without influence and creates precise binary tree structures quickly and efficiently.