SAS Advanced Predictive Modeling

Exam Content Guide

Neural Networks - 20%

• Describe key concepts underlying neural networks
• Use two architectures offered by the Neural Network node to model either linear or non-linear input-output relationships
• Use optimization methods offered by the SAS Enterprise Miner Neural Network node to efficiently search the parameter space in a neural network
• Construct custom network architectures by using the NEURAL procedure (PROC Neural)
• Based upon statistical considerations, use either time delayed neural networks, surrogate models to augment neural networks
• Use the HP Neural Node to perform high-speed training of a neural network

Logistic Regression - 30%

• Score new data sets using the LOGISTIC and PLM procedures
• Identify the potential challenges when preparing input data for a model
• Use the DATA step to manipulate data with loops, arrays, conditional statements and functions
• Improve the predictive power of categorical inputs
• Screen variables for irrelevance and non-linear association using the CORR procedure
• Screen variables for non-linearity using empirical logit plots
• Apply the principles of honest assessment to model performance measurement
• Assess classifier performance using the confusion matrix
• Model selection and validation using training and validation data
• Create and interpret graphs (ROC, lift, and gains charts) for model comparison and selection
• Establish effective decision cut-off values for scoring

Predictive Analytics on Big Data - 40%

• Build and interpret a cluster analysis in SAS Visual Statistics
• Explain SAS high-performance computing
• Perform principal component analysis
• Analyze categorical targets using logistic regression in SAS Visual Statistics
• Analyze categorical targets using decision trees in SAS Visual Statistics
• Analyze categorical targets using decision trees in PROC IMSTAT
• Analyze categorical targets using logistic regression in PROC IMSTAT
• Build random forest models with PROC IMSTAT
• Analyze interval targets with SAS Visual Statistics
• Analyze interval targets with PROC IMSTAT
• Analyze zero inflated models with HPGLM in Enterprise Miner

Open Source Models in SAS - 10%

• Incorporate an existing R program into SAS Enterprise Miner
• Incorporate an existing Python program into SAS Enterprise Miner