Predictive Modeling with Logistic Regression using SAS

This module provides an overview of the course and covers logistical details such as learner prerequisites, accessing SAS software, setting up data, and utilizing forums for assistance. Participants will also gain insights into the course structure and expectations.

Participants will gain a solid understanding of predictive modeling, including the basic steps, applications, and challenges associated with data and analytical aspects. The module includes practical demos, practices, and questions to reinforce learning.

This module focuses on fitting the logistic regression model, covering topics such as model interpretation, estimation, scoring new cases, correcting for oversampling, and assessing model performance. Participants will engage in hands-on practice to reinforce their understanding of model fitting.

Participants will learn about preparing input variables by addressing missing data, handling categorical inputs, collapsing categories, reducing redundancy, and variable clustering methods. Practical exercises and questions will reinforce the learning process.

This module delves further into preparing input variables, covering topics such as detecting nonlinear relationships, variable screening, subset selection methods, detecting interactions, and model selection. Participants will engage in practical exercises and questions to reinforce their understanding.

Participants will learn to measure model performance through assessing fit versus complexity, using confusion matrices, ROC curves, gains charts, profit matrices, K-S statistic, and comparing multiple models. Practical exercises and questions will reinforce the learning process.

This module provides a practice exam for statistical business analysis using SAS®9, focusing on regression and modeling. Participants can assess their understanding and readiness for SAS certification through this practice exam.