Introduction to AnswerTree

Target Audience:
This course will appeal to those wishing to find a means of targeting sub-groups of a population and who work primarily with categorical data, but who may have continuous data. Typical applications include: identifying those most likely to respond positively to a mailing, those who may prove to be greater credit risks or those customers who are likely to churn.

Prerequisites:
You must be PC literate and have a solid understanding of basic statistical concepts (including measures of central tendency, dispersion and cross tabulation tables). You must also know what data is available to you and what you will be trying to achieve using Answertree.

Overview:
The course begins with an overview of tree-based methods and the four methods that are incorporated into Answertree. There is a general introduction to the features of AnswerTree, followed by a detailed look at each of the methods (CHAID, Exhaustive CHAID, C&RT QUEST).

Objectives:
By the end of the course you will have learned:
• The assumptions and concepts underlying Tree-Based Segmentation
• To use all essential features of AnswerTree, as well as selected advanced options
• To control the operation of AnswerTree and the generation of output
• To interpret the output from AnswerTree and draw appropriate conclusions

Duration: 1 day

Course Content:
Initially the course provides a background to AnswerTree and the different methods available. The course will then proceed logically through the following topics:

Overview of Features in AnswerTree
• What method to use and when
• Loading data from SPSS
• Interpreting gains tables
• Interpreting risk summaries
• Creating rules in SPSS Syntax or SQL format

CHAID and Exhaustive CHAID Method
• Applications, aims and assumptions
• Interpretation of output
• Building the segmentation tree
• Tree validation and incorporating Profit & Costs

Classification & Regression Tree (C&RT) Method
• Understanding the principles behind the C&RT method
• Dealing with Gains Tables and Risks Summaries
• Running the Analysis with categorical and continuous dependent variables
QUEST method

Recommendations, tips and efficiency.