Data Mining Cookbook: Introduction to Effective Predictive Modeling
Business Knowledge Series course
Duration: 2.0 days
Course fee:
EPTO units: .
CEUs: 1.2
| |
This course is not currently scheduled.
Presented by C. Olivia Parr-Rud, Data Mining Strategist Consultant, OLIVIAGroup, and author of
The Data Mining Cookbook
This course teaches statisticians and modelers how to assimilate business knowledge into predictive model development. Several methodologies are discussed and compared, with the emphasis is on how to meet a company's business objective through a thorough understanding of how to translate the objective into a case for predictive modeling.
Learn how to
- define the modeling objective
- gather and prepare the data for the model
- select and transform variables to prepare for modeling
- process and evaluate the model
- validate, implement, and maintain the model.
Who should attend
Analysts, statisticians, marketers, and programmers
Expand All
Collapse All
Print version
Prerequisites
Before attending this course, it is recommended that you have some familiarity with database marketing. Proficiency with Base SAS is highly recommended. Basic understanding of SAS DATA steps is required. A general understanding of statistics is helpful.
Course Contents
Defining the Objective
- discuss the model objective as it relates to the overall company or business goals
- discuss a variety of objectives related to different industries
- introduce descriptive and predictive statistical methods
Gathering the Data
- gain a thorough understanding of the many data types and sources
- learn rules for obtaining the best data for each specific modeling project
- learn how to create a modeling data set
- introduce the case study
Preparing the Data for Modeling
- understand the importance of accurate data
- learn to create a modeling data set
- learn sampling techniques
- clean and prepare the data for modeling
Selecting and Transforming the Variables
- learn how to define the objective function in business terms that relate to the company objective
- understand some important techniques in selecting and transforming variables
- learn efficient methods for transforming large numbers of input variables
- learn methods for selecting the final candidate variables
Processing and Evaluating the Model
- understand the characteristics and benefits of logistic regression
- learn some of the different options within logistic regression
- practice running and evaluating the model
Validating the Model
- learn how to build a gains table and lift chart
- understand how to implement bootstrapping validation
- learn to create tables to emphasize the model's effect on any variable
Implementing and Maintaining the Model
- learn how to prepare data for scoring and implementation
- learn how to audit data to insure valid results
- understand the key components and benefits of a model log
Software
This course addresses SAS/STAT. This course also touches on SAS/GRAPH software.
Course Materials
Students receive a hardcopy of the course notes and, in some courses, can choose to take home a copy of the course data.
You also receive a copy of C. Olivia Parr-Rud's book, The Data Mining Cookbook.
Share Your Thoughts
Are there additional topics you'd like for this course to address?
Would you like for this course to be offered at another training facility?
Let us know by adding to our
Interest List.
Course fee and EPTO units will differ for on-site training.
This page was created using SAS software.