SAS^® 9.3

Features

Data management

• Enhanced manageability for improved performance
  • Version and source control, including archival, differencing and rollback, including plug-ins for third-party version control products CVS and SubVersion.
  • Job status and performance reporting for tracking metrics such as CPU utilization, memory, I/O, number of jobs, number of records, etc., in a historical view.
  • Automated job deployment allows use of common scripting languages to deploy SAS Data Integration Studio batch jobs in an automated fashion.
  • New command-line job deployment for deploying single and multiple jobs.
  • Enhanced SAS code import capabilities enable current SAS users to import their SAS jobs into SAS Data Integration Studio. Includes improved logging and error checking.
• In-database processing (ELT pushdown)
  • Optimized SQL set transforms.
  • Support for Oracle bulk loader.
  • Import capabilities for user-defined functions that enable in-database scoring for analytics.
• Enhanced performance for loading Oracle data
  • Provides native support for bulk loading.
  • Includes an easy-to-use GUI for integration.
  • Provides support for flexible loading options, including direct path and load to partition.
  • Includes the ability to drop or recreate indices.
• New ETL transforms
  • New transforms generate SAS code that is very efficient with high performance.
  • New transforms include: Type 1 SCD supports both MERGE and HASH techniques, Table Differencing and enhancements for Type 2 SCD loaders.
  • New Compare Tables transform compares two data sources and detects changes in data.
• Data access
  • Support for new data sources: Teradata, Aster Data, EMC/Greenplum and Sybase IQ.
  • Native support for SQL-based processing. Support for bulk-load utilities.
  • Push-down processing supports in-database processing.

Data quality

• SAS^® Enterprise Data Integration Server includes DataFlux^® Data Management Platform for enhanced data quality
  • Includes integrated profiling, exploration and entity resolution capabilities.
  • Provides powerful process orchestration layer for jobs, SAS code and SQL.
  • Creates the basis for a master data management strategy.
• Master Data Management (DataFlux^® qMDM)
  • Provides support for master hub of trusted entity data.
  • Supports a data-quality-driven approach for mastering key business entities.
  • Provides data models for key business domains.
  • Provides governance and data steward services that support inspection, root-cause analysis and the correction of data and propagation of corrections across the IT environment.

Analytics

• SAS/STAT®
  • Finite mixture models.
  • Random effects for Bayesian models.
  • Cox regression for sample survey data.
  • Updates to mixed models for data with large numbers of observations.
  • Shared frailty models.
  • Fully conditional specification method for multiple imputation.
• SAS/ETS®
  • Simulate realizations or estimate parameters of multivariate distributions by using the copula approach.
  • Powerful and versatile tools for state space modeling and forecasting of time series and longitudinal data.
  • Data access engine SASEXCCM for the CRSP/COMPUSTAT Merged Database.
  • New graphs such as forecast plots, periodogram for an error series, and spectral density estimation plots.
  • New features for modeling the size of losses or insurance claims.
• SAS/QC®
  • Monitor multivariate process variation over time in order to determine whether a process is stable or to detect and diagnose changes in a stable process.
  • Fit Gumbel, inverse Gaussian, generalized Pareto, power function, and Rayleigh distributions.
  • Fit parametric models for recurrent events data and constructs probability plots for models based on a three-parameter Weibull distribution for lifetime data.
• SAS/IML®
  • Call SAS procedures and DATA steps.
  • Provides an interface to the R statistical programming language.
  • New functions and subroutines.
• SAS® Enterprise Miner™
  • Provides new standardized Web Infrastructure Platform mid-tier services for data integration, BI, SAS solutions and analytics.
  • New survival data mining analysis predicts when an event will happen, not just if it will happen (e.g. predicts event probability for time intervals for each customer).
  • New rate-making capability for insurance predicts optimal insurance premium for individuals based on attributes known at application time.
  • Time Series Data Mining node (experimental) applies data mining techniques to transactional, time-stamped data.
  • Support Vector Machines node (experimental) provides a supervised machine learning method for prediction and classification.
  • New interactive tree application features enable users to select optimal tree based on assessment on validation data. Other evaluation plots are automatically updated based on the selection of a sub-tree.
  • Least Angle Regression node is now supported for both interval and binary targets.
  • SAS Rapid Predictive Modeler adds the ability to score the training data with an option to save the scored data set.
  • Improved speed for large data sets.
  • Procedures optimized for utility file usage.
  • SAS code optimized for efficient file usage.
• SAS® Forecast Server
  • Automatic time series combination (ensemble) models allow users to combine different forecasts to produce another forecast.
  • Ability to define customized time intervals.
  • Rolling simulation lets you interactively simulate a forecasting process.
  • Temporal reconciliation enables users to reconcile forecasts across time dimensions (e.g., daily, weekly, monthly, yearly). (Included with the SAS High-Performance Forecasting engine.)
  • Administrators can now assign role-based access to the SAS Forecast Studio GUI.
  • Export model selection list enables modelers to export models to a repository using point and click.
  • Auxiliary data set support.
  • Increased scalability for SAS Forecast Studio.
  • Exploits multikernel machines for faster response times.
  • SAS Forecast Server is integrated with the SAP APO Demand Planning module to provide SAP users with access to a superior forecasting engine and automatic forecasting capabilities.
• SAS® Model Manager
  • Analytic workflow management. Using SAS Workflow Studio, users can define a workflow process for each model based on their needs using notifications, timers, etc.
  • Seamless integration of R models with the ability to register and manage R models in SAS Model Manager.
  • Ability to perform champion/challenger side-by-side comparisons between SAS and R models to see which model performs best for a specific need.
  • Model retraining enables existing models created with SAS Enterprise Miner to be retrained with up-to-date data to generate a new set of modeling results.
  • Improved performance-monitoring dashboards. Update all reports or update reports for projects that have new performance data.
• SAS/OR® and SAS^® Simulation Studio
  • Optimization
    • New network simplex linear programming algorithm.
    • Crossover feature for interior-point linear programming solver (experimental).
    • New multistart capability for nonlinear programming solvers.
    • Performance improvements for all optimization solvers (linear, mixed-integer, quadratic and nonlinear).
  • Simulation
    • Automatic input distribution fitting using JMP with SAS Simulation Studio.
    • Use data to drive models – empirical distributions, nonhomogeneous Poisson processes.
    • Improved usability to simplify navigation of large models.
    • Support for Windows 64 and remote SAS servers.

Text analytics

• SAS® Text Miner
  • Enhanced language support (now supporting 27 languages + dialects).
  • New text import capabilities for sourcing file system and Web data.
  • Reuse synonym data sets and other usability enhancements.
• SAS® Enterprise Content Categorization
  • Wikipedia integration to support the generation of taxonomies & add-on industry-specific taxonomies.
  • New graphical reports enable users to understand the precision and accuracy of the classification process.
  • New operators, including co-reference and pronoun resolution.
• SAS® Sentiment Analysis
  • Enhanced language support (now supporting 27 languages + dialects).
  • Enhanced rule-building, editing, verification and operators.
  • Interactivity makes it easier for subject-matter experts to add new concepts.

Scalability and high performance

• SAS® Analytics Accelerator for Teradata (new product)
  • Enables in-database data discovery and predictive modeling.
  • Allows a key set of SAS statistical and analytical procedures used during data discovery, summarization and predictive modeling to be executed directly within a Teradata database or Teradata data warehouse.
  • Includes SAS/STAT procedures: CORR, CANCORR, FACTOR, PRINCOMP, REG, SCORE and VARCLUS.
  • Includes SAS/ETS^® procedure: TIMESERIES.
  • Includes SAS Enterprise Miner procedures: DMDB, DMINE and DMREG.
• SAS® Grid Manager
  • SAS Stored Process Server, SAS OLAP Server and SAS Workspace Server now support processing in a grid environment.