Sample Size Computations and Power Analysis with the SAS System Abstract
SAS Meets Big Iron: High Performance Computing in SAS
Analytic Procedures"
Robert A. Cohen, SUGI Proceedings, 2002.
Abstract
Version 9 targets the heavy-duty analytic procedures in SAS for high performance computing enhancements. These enhancements encompass both algorithmic improvements and modifications to exploit multiprocessor hardware. This paper provides a survey of this development and the performance gains obtained in several procedures in SAS/STAT and Enterpreise Miner. Some general scalability issues are discussed and their practical implications in providing high performance scalable implementations in the following SAS facilities are addressed:
- linear regression
- analysis of variance
- local regression
- robust regression
- logistic regression