New in JMP® 18
See the newest capabilities in JMP, JMP Live, JMP Pro, and JMP Clinical
Configurable Data Connectors
JMP gives you easy access to your data from a variety of sources. With the additional data connectors now available with JMP 18, even more data is available to enrich your analyses. These data connectors are configurable, for more options to customize the details of your connections.
- Use customizable built-in configurations for various database systems with examples to reference in the scripting index.
- Import data using the streamlined PI Server Import. Multiple attributes can be easily imported into a single data table, multiple rows of data within a given time interval can be collapsed into a single row, and data can be sampled at a specified interval.
- Customize the details of how JMP connects with a data source using configurable data connectors that support ODBC connections.
- Easily share connectors will colleagues to facilitate their work.
- Publish and refresh data to JMP Live using configurations that are compatible with both Mac and Windows.
Platform Presets
When you have customized a report to your liking (including colors, axis settings, red triangle options), you can capture those customizations for use with future launches of the same platform. JMP includes a set of popular platform presets, or you can save your own preferred configuration as a preset for any data table or set of columns.
- Save often-used report output customizations for efficient analysis and sharing.
- Create presentation- or publication-ready outputs to share with colleagues.
- Provide guidance on how to perform common analyses in reports using presets that have the correct red triangle options pre-selected.
- Assist new users in setting up platforms used frequently in your organization.
Model Profiler Enhancements
The Prediction Profiler is a staple in the analytic workflow for many data analysts, allowing practical, interactive model exploration. Now you have more options for using profilers to understand your predictive models.
- Show prediction intervals for predicted (individual) values, as well as confidence intervals for predicted means.
- Show data points and overlaid interaction traces on profilers.
- Save prediction and interval formulas in one step in model fitting platforms. This feature makes it easier to plot confidence and prediction intervals using Graph >Profiler.
- Use the simpler, easier launch dialog in all profiler platforms under the Graph menu.
- View and update constraints in the profiler more easily.
- Use K Nearest Neighbors-based extrapolation control.
- Publish profilers with prediction intervals and interactive profilers for more model types in JMP Live.
New in JMP Pro
Tackle even more complex and challenging analytic problems using the latest data science techniques in JMP Pro, including advanced predictive modeling and machine learning with extended analytic capabilities.
- New methods for Direct FPCA in FDE, including Multivariate Curve Resolution and Penalized SVD. Peak detection is also available for chromatography and spectrometry profiles.
- Enhanced GLMM is now more full-featured, supporting additional distributions and covariance structures.
- More Genomics capabilities, a new marker relatedness platform, and improvements to marker statistics, marker simulation, and multivariate embedding (UMAP/t-SNE) platforms.
- Models that include random effects can now be specified in Fit Response Screening.
- Enhancements to SEM including new model shortcuts, platform-wide bootstrapping, and robust inference.
Improved Python Integration
For data analysts, scientists, and engineers using Python, the new JMP integration simplifies the programming to keep the focus on the analysis. JMP 18 includes a single, supported version of Python, an extensive set of standard Python libraries, a more robust connection between JMP and Python data structures, and an improved Python coding experience within JMP.
- Run the integrated Python version in JMP using the new Python editor for a simple but familiar Python programming experience.
- Access to JMP data tables in memory from Python directly; easy access to Python data from JMP.
- Run JSL from Python.
- Continue using JMP with other software, such as SAS, R, and MATLAB, using openly available Python packages.
New in JMP Clinical
An enhanced suite of tools in JMP Clinical 18 gives you expanded options for interactively exploring your clinical trial data, more easily filtering, and quickly screening for safety issues.
- New comprehensive Drug-Induced Liver Injury (DILI) report based on FDA guidelines for identifying subjects in a trial that have possible liver injury.
- Clinical Data Mapping Tool for converting raw data to STDM data standards.
- Redeveloped Risk-Based Monitoring report with updated visualization of the risk metric scores
- A GUI-driven Subpopulation Builder to specify search and filter criteria to identify subpopulations of trial subjects that meet ad hoc or predefined criteria.
- Clinical Recoder to recode study variables and to allow combining or splitting levels for a variable.
New in JMP Live
JMP Live 18 offers expanded support for the customer/supplier model and simplified user management that allow scientists and engineers in your organization to drive innovation and empower data-driven decisions more efficiently.
- Direct user and group synchronization with Microsoft Active Directory.
- Tagging users in comments (using @ mention).
- Various file types can be uploaded to refresh a data table on JMP Live.
- Permission configurations at the folder level.
- Enhanced post sorting by different criteria.
- Option to delete users, including their personal identifiable information (PII).
- New Profiler platforms.
- Easier navigation with enhanced UI elements, including side rails and a new folder view.
Additional Items
- Loading JMP add-ins in JMP sessions.
- Support for multiple local data filters.
- Options for making report changes with red triangle menus in distributions.
- New confidence intervals.
- Enhanced pattern filling and data/time patterns.
- Improved authentication for data access.
- History for reports or all content in a space.
- Directory access for refresh scripts.
- Support for a private key in Snowflake authentication.
Other Key Features
Data Access, Prep, and Exploration
Graph Builder Updates
View graphs as a grid, with all graphs on independent scales. New Label option on the Points element. New control of the color and style of overlay variable levels.
Collapsible Annotations
New text annotations in reports can be collapsed or expanded to hide or reveal additional information.
Columns Manager
New platform to view and manage column properties. Enhanced Columns Manager will replace Columns Viewer in JMP 19.
Data Table Updates
- Enhancements to column width options.
- New option for displaying statistics in the column header.
- Character columns can be marked compact, which, in most cases, uses less space in memory and storage. Tables with compact character columns cannot be opened in JMP 17 and earlier.
- Column headers can now be colored. This color is visible in the table header and also in column lists throughout JMP.
- New Insert Rows command to quickly insert rows at the current selection.
Group Platforms
Creates an outline box and red triangle menu to operate on a group of related platforms. Useful for anyone who uses groups or launches with multiple Ys. Simplifies simultaneous interaction with multiple platforms.
PI Server Import Enhancements
Multi-attribute import; stacking functionality to reshape imported data; "Update" table script to import and concatenate data to the existing data table; and "Change Attributes" table script to modify the data table by adding columns instead of rows.
Wide Transforms
Applies an arbitrary formula to a column (or columns) and creates a new formula column, or writes the results in place of the original column.
Additional Items
- Numbers in scientific notation displayed as powers of 10.
- New Contour Platform Level customization dialog.
- UI support to export JMP data as V8 SAS transport file format.
- Text in many prompt dialogs now readable via accessibility technology.
Design of Experiments, Reliability, and Process and Quality Engineering
New Constant Stress ALT Platform
Design and compare accelerated life tests with up to three accelerating factors. Replaces the legacy Accelerated Life Test Design platform.
Degradation Platform Updates
Fit built-in nonlinear degradation models and linear degradation models through scripting, without initial values. More easily select the appropriate model fitting type to launch the desired analysis.
Easy DOE Updates
Improvements to adding, editing, and reordering factors and responses, improvements to the analysis of Definitive Screening Designs, respect for effect heredity, prospective confidence intervals, enhanced Load Design, and other improvements.
New Fatigue Model Platform
Analyze fatigue data, also known as S-N (strain or stress versus number of cycles) curve modeling.
Orthogonal Mixed-Level Designs
Screening designs for continuous and two-level categorical factors where the continuous factors can be at three levels as an alternative to DSDs.
Parametric Survival Updates
Confidence intervals are now in the residual probability plot. New Hazard Profiler allows inspection of hazard function, given the values of other factors.
Reliability Block Diagram Updates
Save a Reliability Block Diagram as a Repairable Systems Simulation Diagram. Export Reliability Block Diagram Component Distribution Settings to a JMP table.
Repairable System Simulation Update
Export Repairable System Simulation Component Distribution Settings to a JMP table.
Short Run Control Charts
Monitor short-run production data with new control charts.
Process Screening Updates
Plot Cp vs. %Measurement Sigma2 with a new graph called Process Potential.
Additional Items
- Fit Life by X updates: Legend line styles and line colors are customizable. Three new scripts (Time Series, Scatter Plot Matrix, and Distribution) provide diagnostics of posterior samples.
- Enhancements to Control Chart dialogs under Quality and Process > Control Chart. Legacy Control Charts will be retired in JMP 19.
- Life distribution is updated with three new scripts (Time Series, Scatter Plot Matrix, and Distribution) that provide diagnostics of posterior samples when using Bayesian estimation.
- MSA Design enhancements: New diagnostics are based on standardized residuals and other platform improvements.
- Tolerance intervals can now be saved to a Spec Limits column property in the Distribution platform.
Statistics, Predictive Modeling, and Data Mining
Fit Curve Updates
Additional peak models (including Pseudo-Voigt, Pearson VII function, Skew Normal function, and Asymmetric Normal); equivalence tests now use Fieller’s method by default; and T2EQ method added for comparing dissolution profiles.
Matched Pairs Update
New options for equivalence testing and analyzing percent differences
Nonparametric Equivalence Test
Updated equivalence testing in Oneway that includes a nonparametric test option.
Precision Recall Curve
Shows the performance of a classifier, plotting the Precision values against the Recall values at a variety of thresholds. PR Curves are now available in many JMP platforms, including Model Screening.
Principal Components
Additional principal components methods: Nonnegative Principal Components and Sparse Components Analysis.
Design Space Profiler Updates
New menu commands Reset Factor Space, Get Midpoints from Profiler, Show Current Profiler Values, and Send Limits to Simulator as Constraints. Send Limits to Simulator offers two new distributions, Normal weighted at 3 Sigma and Normal weighted at 2 Sigma. Make and Connect Random Table embeds scatter plots directly into the report.
Distribution Enhancements
Goodness of fit test for binomial distribution fits, as well as the ability to fit the zero-inflated binomial, the zero-inflated SHASH, and the exponentially modified Gaussian distributions. Ability to switch the CDF plot Y-axis from linear scale to the desired probability scale.
Additional Items
- Connected letters report updates: Non-parametric multiple comparison is now in Oneway.
- Wilcoxon Kruskal-Wallis, Median, Friedman Rank, or Van der Waerden nonparametric tests of groups now provide one-sided test p-values.
- One-sided Dunnett’s multiple comparison with a control are now available in the Oneway platform.
- Cohen’s d effect size measurement is now reported in the pooled t-test report in the Oneway platform.
- Jonckheere Terpstra Test is now available in the Oneway and Contingency platforms. This test is often used for dose-response testing.
- Effects in the Standard Least Squares personality of the Fit Model platform can be excluded using controls in the Effect Summary panel.
- More flexibility in specifying knot points for knotted splines in the Fit Model platform.
- Fit Mixed update adds VIF to the Fixed Effects Parameter Estimates table.
- Nonlinear platform updates: Usability improvements implemented. Platform now provides Profile Likelihood and Profile Likelihood Contour for Likelihood-Based Model Fitting.
- Randomized Complete Block Design updated to support Post-Hoc Comparisons for Friedman Rank Test.
- The Means Differences report in Response Screening now contains unadjusted confidence intervals and Benjamini-Yekutieli FDR adjusted confidence intervals for the differences in means.
- The Simulation Experiment now respects linear constraints that are incorporated into the Prediction Profiler.
Scripting, Automation, and Workflow
Workflow Builder Update
Conditional and looping support at the step level that makes it possible to handle multiple tables in the same workflow and to execute steps based on JMP analysis results or user input.
LRT Menu Search
Search in red triangle menu or context menu at the top of the menu for easy skimming and selection.
Help Menu Enhancements
Simplification of the Help menu, sample index improvements, and guidance for new users.
Additional Items
- Updates to JSL functionality:
- Support for codes to labels and labels to code.
- Support for maximum JMP version indicator.
- Scripting-only feature providing a new Where function that provides a concise way to filter columns, lists, and matrices.
- Updated J function OSI examples with additional examples that illustrate efficient implicit iterations, for both numerical computation and non-numerical purposes.
- User access to automated initial values calculation for linear models and built-in nonlinear models.
- General support for representing colors in different color spaces other than sRGB, including support for Display-P3 color space.
- On Windows, the Save dialog now supports the XML file type.
