Description
This course covers predictive modeling using SAS/STAT software with emphasis on the LOGISTIC procedure. This course also discusses selecting variables and interactions, recoding categorical variables based on the smooth weight of evidence, assessing models, treating missing values, and using efficiency techniques for massive data sets. You learn to use logistic regression to model an individual’s behavior as a function of known inputs, create effect plots and odds ratio plots, handle missing data values, and tackle multicollinearity in your predictors. You also learn to assess model performance and compare models.
What you will learn
Course Overview and Logistics
Understanding Predictive Modeling
In this module, you review the fundamentals of predictive modeling. Then you explore the business scenario data that is used throughout the course. Finally, you learn about common analytical challenges that you might encounter as a modeler.
Fitting the Model
In this module, you investigate the concepts behind the logistic regression model. Then you learn to use the LOGISTIC procedure to fit a logistic regression model. Finally, you learn how to score new cases and adjust the model for oversampling.
Preparing the Input Variables, Part 1
In this module, you learn how to deal with common problems with your predictor variables such as missing values, categorical predictors with many levels, a high number of redundant predictors, and nonlinear relationships with the response variable.
Preparing the Input Variables, Part 2
In this module, you learn how to select the most predictive variables to use in your model.
