turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- What ML algorithms are part of base SAS

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

08-28-2015 03:07 PM

Am I correct to say that simple learning algorithms like linear and logistic regression can be done in sas enterprise guide using proc reg, but advanced algorithms like SVM and neural networks require enterprise miner?

Accepted Solutions

Solution

08-28-2015
03:35 PM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

08-28-2015 03:35 PM

Yes, neural networks, random forests, SVM are limited to enterprise miner. You need SAS/STAT to be able to use proc reg but I think that it comes standard in most packages. You also have access to clustering algorithms (k-means,hierarchical agglomerative clustering,dendrograms) through procs such as varclus, cluster, and tree.

Best,

Daniel

All Replies

Solution

08-28-2015
03:35 PM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

08-28-2015 03:35 PM

Yes, neural networks, random forests, SVM are limited to enterprise miner. You need SAS/STAT to be able to use proc reg but I think that it comes standard in most packages. You also have access to clustering algorithms (k-means,hierarchical agglomerative clustering,dendrograms) through procs such as varclus, cluster, and tree.

Best,

Daniel

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

08-28-2015 03:36 PM

It sounds like you are including SAS/STAT in your definition of "Base SAS". SAS/STAT has logistic regression, discriminate analysis, clustering, variable selection, GAM models, and the new HPSPLIT procedure for regression trees. I don't think it is fair to classify these as "simple" algorithms, but many of them are "traditional." For an overview of the algorithms in EM, see http://support.sas.com/resources/papers/proceedings14/SAS313-2014.pdf