SAS DELIVERS JMP V.4.0 FOR STATISTICAL DISCOVERY

SAS Institute, a market leader in integrated data warehousing and decision support, demonstrated JMP Statistical Discovery Software, Version 4.

Demonstrated at both the Drug Information Association Conference in San Diego, and the Quality Expo Detroit, JMP software is an award-winning statistical data analysis solution for Windows and Macintosh. Its approach to Design of Experiments (DOE) with Version 4 allows users to simply describe their manufacturing problem (factors, interactions, budget constraints, and goals) and JMP constructs a tailored design using its new Custom Designer.

For more than a decade, scientists and engineers have used JMP to improve process quality, but now with Version 4, they can also optimize their processes in fewer runs. JMP new Custom Designer allows manufacturers to create a design for any situation. If classical design meets the requirements, JMP's Custom Designer constructs a design equivalent to the classical design. If not, it will find the best design for the situation.

"The new and improved JMP 4 is the easiest to learn and the best tool we've found for classroom and research use by applied scientists," said Dr. Andy Mauromoustakos, associate professor at the University of Arkansas and Version 4 beta test site. "It helps researchers visualize their data, aids their understanding of models being fitted, and provides the necessary tools to easily and efficiently communicate their results."

"Knowing that SAS welcomes their feedback, our customers continuously drive us to put new features in the software as well as suggest ideas for improving the interface," said John Sall, SAS senior vice president and originator of JMP. "With JMP 4, we've worked hard to add enhanced functionality without clutter or complication."

New users can get up and running quickly using the JMP Starter menu, offering the same capabilities as the standard JMP menu bar but also providing brief descriptions of each task. JMP 4 allows a user to program repetitive analyses with a new JMP Scripting Language (JSL) to produce scripts that automate analysis. JMP power users may want to use JSL to write custom programs to manipulate data in complex ways, including full matrix algebra support.

"There's no longer a need to find a tabled design and struggle to match the problem to it," says Bradley Jones, principal developer of JMP's Custom Designer. "Many years of working with engineers on DOE studies has resulted in this method of custom design which enables researchers to reduce the number of trial runs necessary to optimize a process. We expect savings for manufacturing users by lowering both their resource costs and production down time." Increased support for DOE includes a new Taguchi Arrays platform for constructing inner and outer arrays as well as calculating signal-to-noise ratios.

In addition to pioneering this new direction in DOE, JMP also includes many classic statistical methods and graphics used by other analytical packages. Its core includes strong data exploration capabilities and an intuitive approach to statistics that lead to valuable data discoveries and ultimately to better decision-making. Statistical support in JMP 4 has been expanded to include methods and diagnostic tools for evaluating time-series data, and restricted maximum likelihood (REML) methods for mixed models. It will also fit a number of distributions including Normal, Lognormal, 2 & 3 parameter, Weibull, Extreme Value, Gamma, Beta, and Exponential.

JMP 4 has been totally rewritten in C++ to more easily maintain and enhance the product in the future. Additional enhancements include an HTML-based help system, documentation presented online as Adobe Acrobat Portable Document Format (PDF) files, access to all ODBC databases, a new Layout facility to customize presentations, and an option to save reports directly to HTML (with graphics automatically embedded in JPEG, PNG or WMF files) for Web or intranet delivery.

About SAS

SAS Institute, one of the world's largest privately held software companies, is a leader in decision support and data warehousing, providing integrated enterprise information-delivery and e-business solutions. Founded in 1976, SAS markets packaged business solutions for vertical industry and departmental applications, as well as an integrated suite of software tools and consulting services. These allow companies to transform the wide variety of data within their organizations into information that business users and researchers need to make better decisions. SAS software and services are used at more than 33,000 business, government, and university sites in over 115 countries. SAS' 1999 revenues totaled $1.02 billion. Please visit SAS' Web site: www.sas.com.