Overview

JMP Pro is a sophisticated predictive analytics software tailored for scientists and engineers. It extends the capabilities of traditional JMP software, enabling users to tackle larger and more intricate analytical challenges using the latest data science techniques, including predictive modeling and machine learning. With JMP Pro, professionals can build robust models to anticipate outcomes related to new customers, processes, or potential risks. The software supports the integration of diverse data types, such as unstructured text from repair logs, engineering reports, and customer feedback, transforming them into actionable insights for more informed decision-making. Additionally, JMP Pro streamlines the process of screening, fitting, and comparing multiple models, allowing users to efficiently identify the best fit for their data and generate scoring code in various programming languages like C, Python, JavaScript, SAS, or SQL.

Features and Capabilities

• Predictive Modeling and Cross-Validation: Utilize a comprehensive set of algorithms to construct and validate models effectively.
• Model Screening and Comparison: Develop multiple candidate models, profile them, and determine the optimal one for specific analytical problems.
• Formula Depot and Score Code: Organize models systematically and save scoring code in formats compatible with SAS, C, Python, JavaScript, or SQL.
• Structural Equation Modeling (SEM): Fit a variety of models, including confirmatory factor analysis, path models, measurement error models, and latent growth curve models.
• Modern Modeling Techniques: Employ advanced methods, such as generalized regression with penalized approaches, to build superior models even when faced with challenging data.
• Functional Data Analysis: Model data that are functions, signals, or series using the Functional Data Explorer (FDE).
• Reliability Block Diagrams: Identify and address weak points in systems to prevent future failures.
• Repairable Systems Simulation: Simulate repair events to understand system downtime, frequency, and associated costs.
• Covering Arrays: Design experiments that maximize the probability of detecting defects while minimizing cost and time.
• Term Selection and Sentiment Analysis: Analyze unstructured data to identify terms associated with responses and explore basic sentiment.
• Mixed Models: Analyze data involving both time and space, where multiple subjects are measured or groups of variables are correlated.
• Uplift Models: Predict consumer segments most likely to respond favorably to an action, facilitating targeted marketing decisions.
• Generalized Linear Mixed Models (GLMM): Fit models with both non-Gaussian response variables and random design effects.