Chapter 12

Considerations for Predictive Modeling

 

12.1 Introduction

12.2 Target Windows and Observation Windows

12.3 Multiple Target Windows

12.4 Overfitting

 

12.1  Introduction

Predictive modeling is a discipline of data mining that plays a very important role. Because voluminous literature can be found concerning the training and tuning of predictive models, we will not consider these topics here. However, because the analysis data mart is a very important success factor for the predictive power of models, we will consider some design details for predictive modeling. In Chapter 20 – Coding for Predictive Modeling, we will deal with the creation of derived variables with high predictive power.

The points that we discuss here mostly concern ...

Get Data Preparation for Analytics Using SAS now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.