SAS

SAS
Name	SAS
Developer	SAS Institute
Released	1976
Latest release	SAS Viya (continuous)
Operating system	Windows, Linux, UNIX
License	Proprietary
Website	sas.com

SAS SAS is a proprietary suite of analytics software for data management, advanced analytics, business intelligence, and predictive modeling. Created by SAS Institute in the 1970s, the system has been used across industries including finance, healthcare, pharmaceuticals, telecommunications, and government. Its ecosystem includes legacy base components and modern cloud-native platforms designed for large-scale analytics, reporting, and regulatory compliance.

History

SAS was originated by a team at North Carolina State University in the early 1970s to analyze agricultural experiments and clinical trials, evolving through collaboration with researchers at International Business Machines Corporation, Cary (North Carolina), and commercial partners. Early milestones include the release of a statistical package in the late 1970s and adoption by universities such as Harvard University and Stanford University for academic research. Through the 1980s and 1990s SAS expanded into corporate markets, competing with vendors like SPSS and SYSTAT, while integrating with enterprise systems from Oracle Corporation and Microsoft Corporation. Regulatory events such as the adoption of Good Clinical Practice standards and legislation like the Health Insurance Portability and Accountability Act of 1996 drove uptake in regulated industries. In the 2010s SAS Institute shifted toward cloud architectures to address competition from open-source projects such as R (programming language) and Python (programming language) and commercial platforms like IBM Watson and Google Cloud Platform.

Overview and editions

The SAS product family includes legacy Base components, modular solutions, and newer platforms such as SAS Viya. Editions and modules have included SAS/Base, SAS/STAT, SAS/GRAPH, SAS/ACCESS, SAS/ETS, and SAS/OR, alongside industry suites for pharmaceutical and banking sectors. Deployments range from single-server installations on Microsoft Windows Server and Red Hat Enterprise Linux to clustered, multi-tenant configurations on Amazon Web Services and Microsoft Azure. Integration points and connectors exist for Apache Hadoop, Teradata, and SAP ERP environments, enabling data interchange with Snowflake (company) and Cloudera distributions.

Features and components

Core components provide data step processing, procedures for statistical analysis, and reporting engines. Notable modules include the SAS/STAT procedures for regression and survival analysis, SAS/GRAPH for visualization, and SAS/IML for matrix computations; connectors such as SAS/ACCESS enable interaction with MySQL, PostgreSQL, and IBM Db2. Enterprise features include job scheduling, metadata management, auditing, and role-based access aligned with standards from ISO/IEC frameworks. Modern additions bring RESTful APIs, microservices, and support for distributed computing frameworks like Kubernetes and Apache Spark for model scoring and batch processing.

Programming language and syntax

The SAS language combines a procedural data step with a large library of PROC procedures for statistical and analytical tasks. Syntax elements include DATA steps, PROC statements, macros, and formats; macro programming enables code generation and parameterization, while the SAS Component Language (SCL) supports custom application development. The language interoperates with external code via PROC PYTHON, PROC IML calling interfaces, and SAS/CONNECT sessions to link with R (programming language) and Python (programming language) ecosystems. Output Delivery System (ODS) controls rendering to formats compatible with Microsoft Excel and Adobe PDF.

Applications and use cases

SAS is widely used in clinical trial analysis for submissions to regulatory agencies such as the U.S. Food and Drug Administration and European Medicines Agency, econometrics and forecasting in central banks including the Federal Reserve System, fraud detection in payment networks tied to Visa Inc. and Mastercard Incorporated, risk modeling in investment banks like Goldman Sachs and JPMorgan Chase, and customer analytics in telecommunications firms such as Verizon Communications and AT&T Inc.. Other domains include supply chain optimization for manufacturers like General Electric and Siemens, and public health surveillance at institutions such as the Centers for Disease Control and Prevention.

Licensing and development

SAS Institute distributes software under proprietary licensing models, offering enterprise agreements, subscription plans, and academic licenses for institutions such as Massachusetts Institute of Technology and University of Oxford. Development is internally managed by SAS Institute with releases timed for feature updates and regulatory requirements; partnerships and certification programs exist for system integrators such as Accenture and Deloitte. Certification tracks and training materials are provided through SAS Institute’s global education services and authorized partners.

Criticism and controversies

Criticism has focused on licensing costs, vendor lock-in, and the steep learning curve compared with open-source alternatives like R (programming language) and Python (programming language). Debates have arisen over interoperability with cloud-native tools from Amazon Web Services and Google Cloud Platform, and legal disputes over competitive practices have involved technology firms and consulting partners. Academic and industry communities have discussed reproducibility and transparency concerns relative to openly accessible libraries and publications such as those in Journal of the American Statistical Association.

Category:Statistical software