Predictive Modeling for Continuous Targets Using IBM SPSS Modeler (v18)
This course (formerly Predicting Continuous Targets Using IBM SPSS Modeler (v16)) provides an overview of how to use IBM SPSS Modeler to predict a target field that describes numeric values. Students will be exposed to rule induction models such as CHAID and C&R Tree. They will also be introduced to traditional statistical models such as Linear Regression. Machine learning models will also be presented. Business use case examples include: predicting the length of subscription (for newspapers, telecommunication, job length, and so forth) and predicting claim amount (insurance).
IBM SPSS Modeler Analysts who have completed the Introduction to IBM SPSS Modeler and Data Mining course who want to become familiar with the modeling techniques available in IBM SPSS Modeler to predict a continuous target.
• Experience using IBM SPSS Modeler, including familiarity with the IBM SPSS Modeler environment, creating streams, importing data (Var. File node), basic data preparation (Type node, Derive node, Select node), reporting (Table node, Data Audit node), and creation of models.
• Introduction to IBM SPSS Modeler and Data Mining (v18)
Please refer to course overview.
1. Introduction to predicting continuous targets
• List three modeling objectives
• List two business questions that involve predicting continuous targets
• Explain the concept of field measurement level and its implications for selecting a modeling technique
• List three types of models to predict continuous targets
• Determine the classification model to use
2. Building your tree interactively
• Explain how CHAID grows a tree
• Explain how C&R Tree grows a tree
• Build CHAID and C&R Tree models interactively
• Evaluate models for continuous targets
• Use the model nugget to score records
3. Building your tree directly
• Customize two options in the CHAID node
• Customize two options in the C&R Tree node
• List one difference between CHAID and C&R Tree
4. Using traditional statistical models
• Explain key concepts for Linear
• Customize one option in the Linear node
• Explain key concepts for Cox
• Customize one option in the Cox node
5. Using machine learning models
• Explain key concepts for Neural Net
• Customize one option in the Neural Net node
