This course covers advanced topics using SAS Enterprise Miner including how to optimize the performance of predictive models beyond the basics. The course continues the development of predictive models that begins in the Applied Analytics Using SAS® Enterprise Miner™ course, for example, by making use of the two-stage modeling node. In addition, some of the newest modeling nodes and latest variable selection methods are covered. Tips for working in an efficient way with SAS Enterprise Miner complete the course.
Zielgruppe / Who should attend
Advanced predictive modelers who use Enterprise Miner
Before attending this course, it is recommended that you
In diesem Kurs wird mit folgenden Software Modulen gearbeitet / This course addresses SAS Enterprise Miner Software
SAS Enterprise Miner Prediction Fundamentals- SAS Enterprise Miner prediction setup
- prediction basics
- constructing a decision tree predictive model
- running the regression node
- training a neural network
- comparing models with summary statistics
Advanced Methods for Unsupervised Dimension Reduction- describe principal components analysis
- describe variable clustering
Advanced Methods for Interval Variable Selection- explain how to use partial least squares regression in SAS Enterprise Miner
- use LAR/LASSO for variable selection
Advanced Methods for Nominal Variable Selection and Model Assessment- implementing categorical input recoding
- creating empirical logit plots
- implementing all subsets regression
Advanced Predictive Models- describe the basics of support vector machines
- use the HP Forest node in SAS Enterprise Miner to fit a forest model
- modeling rare events
- use the Rule Induction node in SAS Enterprise Miner
Multiple Target Prediction- appraising model performance
- defining a generalized profit matrix
- creating generalized assessment plots
- using the Two-Stage Model node
- constructing component models
Tips and Tricks with SAS Enterprise Miner- using the Open Source Integration node
- reusing metadata
- importing and use of external models (self-study)