70--SOURCES SOUGHT

This Sources Sought will be used for market research purposes only. The Government is conducting market research to assist in determining the appropriate acquisition strategy. The Government is seeking responses from businesses capable of providing software license renewal or an alternate software program with equivalent specifications. All responsible sources may submit a capability statement, proposal, or quotation in response to any solicitation that results subsequent to this sources sought notice, which shall be considered by the agency. Seeking companies capable of providing software with the EXACT specifications and capabilities of the JMP Academic Suite (Standard Edition and PRO, simulation studio, data exploration and ANOVA and regression) program which includes, but is not limited to: Must combine sophisticated algorithms to explore, analyze data, fit and compare statistical metamodels for predictive analytics with unique methods. In addition it shall provide the facilities to build custom dashboards for advance decision support. Key must features include: Comprehensive facilities for univariate linear and nonlinear regression, the more useful multivariate approaches for exploration, dimension reduction and modeling, and for the analysis of time series and categorical data. A rich set of sophisticated algorithms for building better models with your data. Allows the user to hold back data consistently to help build models whose predictions will generalize well. Software shall be able to easily compare competing models. Shall have a profiler making it simple to compare and contrast the interpretability and results of various fits. The Profiler shall also allows you to find settings to optimize your outputs, and Monte Carlo simulations help you assess how variation in the inputs will be transmitted into the outputs. Provide a partition platform enables analyst to find cuts or groupings within your inputs variables that can best predict the variation in an output. Provide a way to split the data by finding an appropriate input and an appropriate grouping or cut-point for this input recursively. Partition trees shall be robust to the presence of missing values, and accommodate any joint effects of inputs directly. Decision trees shall have the ability to use decision trees, bootstrap forests or boosted trees. Shall provide customized analytical applications to build dashboards that can analyze and act on data in the context any problem as needed. Software shall support statistical discovery with a unique combination of interactive graphics and wealth of built-in statistics. Shall serve as an analytic hub, allowing the user to quickly and easily build and share custom applications. Shall be able to build and distribute analytic applications while leveraging a drag-and-drop interface to visually design windows with buttons, lists, graphs and other objects. Shall be able to integrate unique scripts to create custom applications. RESPONSES MUST INCLUDE SPECIFICATIONS FOR SUGGESTED SOFTWARE. Responses due on November 16, 2014.